Bioengineering

12 pages, 1226 KiB

Open AccessReview

Fabrication of Artificial Nerve Conduits Used in a Long Nerve Gap: Current Reviews and Future Studies

by Ryosuke Kakinoki, Yukiko Hara, Koichi Yoshimoto, Yukitoshi Kaizawa, Kazuhiko Hashimoto, Hiroki Tanaka, Takaya Kobayashi, Kazuhiro Ohtani, Takashi Noguchi, Ryosuke Ikeguchi, Masao Akagi and Koji Goto

Bioengineering 2024, 11(4), 409; https://doi.org/10.3390/bioengineering11040409 - 22 Apr 2024

Viewed by 1504

Abstract

There are many commercially available artificial nerve conduits, used mostly to repair short gaps in sensory nerves. The stages of nerve regeneration in a nerve conduit are fibrin matrix formation between the nerve stumps joined to the conduit, capillary extension and Schwann cell [...] Read more.

There are many commercially available artificial nerve conduits, used mostly to repair short gaps in sensory nerves. The stages of nerve regeneration in a nerve conduit are fibrin matrix formation between the nerve stumps joined to the conduit, capillary extension and Schwann cell migration from both nerve stumps, and, finally, axon extension from the proximal nerve stump. Artificial nerves connecting transected nerve stumps with a long interstump gap should be biodegradable, soft and pliable; have the ability to maintain an intrachamber fibrin matrix structure that allows capillary invasion of the tubular lumen, inhibition of scar tissue invasion and leakage of intratubular neurochemical factors from the chamber; and be able to accommodate cells that produce neurochemical factors that promote nerve regeneration. Here, we describe current progress in the development of artificial nerve conduits and the future studies needed to create nerve conduits, the nerve regeneration of which is compatible with that of an autologous nerve graft transplanted over a long nerve gap. Full article

(This article belongs to the Special Issue Innovations in Nerve Regeneration)

► Show Figures

Figure 1

Figure 1
Commercially available artificial nerves. Left: Renerve® (Nipro, Osaka, Japan); right: Nerbridge® (Toyobo, Osaka, Japan). The conduit chambers are occupied by collagen materials in both nerve conduits. Full article ">Figure 2
Schematic diagrams of three kinds of nerve conduit tubes: VCT (vessel-containing tube); ET (empty tube); and LVCT (ligated vessel-containing tube). When a vessel-containing tube (VCT) was created, a myocutaneous flap supplied by the sural vessels was elevated from the lower hind limb and turned proximally. The sural vessel pedicle was inserted into the chamber space through a longitudinal slit of the silicone tube, which was sealed with silicone liquid after the pedicle insertion. Each tube was transplanted between the transected sciatic nerve stumps, leaving a 10 mm gap. In LVCTs, the sural vessels inserted into the conduits were ligated. Full article ">Figure 3
Schematic diagrams of the I-group (a sural vessel pedicle was passed through the chamber space of a PGA conduit) and the E-group (a sural vessel pedicle was attached to the outer surface of a PGA conduit). The interstump gap was set at 5 mm. Full article ">Figure 4
Schematic diagrams of a PGA conduit containing two barrels of DABLs seeded with 3 × 106 BMSCs (conduit group) and a 20 mm long autologous nerve graft (autograft group). In the conduit group, the sural vessel pedicle was attached to the outer surface of the PGA conduit. Full article ">Figure 5
Creation of an artificial nerve conduit with a long interstump gap. Full article ">Figure 6
In the PGA group, two pieces of 20 mm long DANBLs (created using chemical surfactants) seeded with 3 × 106 BMSCs was transplanted in a 23 mm long PGA conduit. A sural vessel pedicle was attached to the outer surface of the conduit. In the silicone group, a 20 mm long DANBL (created using the thermal method) seeded with 1 × 107 BMSCs was transplanted in a 23 mm long silicone conduit. A sural vessel pedicle was passed through the conduit lumen. The auto group consisted of a 20 mm long autologous nerve graft model. Full article ">

12 pages, 1150 KiB

Open AccessArticle

How the Effect of Virtual Reality on Cognitive Functioning Is Modulated by Gender Differences

by Stefania Righi, Gioele Gavazzi, Viola Benedetti, Giulia Raineri and Maria Pia Viggiano

Bioengineering 2024, 11(4), 408; https://doi.org/10.3390/bioengineering11040408 - 21 Apr 2024

Cited by 1 | Viewed by 1376

Abstract

Virtual reality (VR) can be a promising tool to simulate reality in various settings but the real impact of this technology on the human mental system is still unclear as to how VR might (if at all) interfere with cognitive functioning. Using a [...] Read more.

Virtual reality (VR) can be a promising tool to simulate reality in various settings but the real impact of this technology on the human mental system is still unclear as to how VR might (if at all) interfere with cognitive functioning. Using a computer, we can concentrate, enter a state of flow, and still maintain control over our surrounding world. Differently, VR is a very immersive experience which could be a challenge for our ability to allocate divided attention to the environment to perform executive functioning tasks. This may also have a different impact on women and men since gender differences in both executive functioning and the immersivity experience have been referred to by the literature. The present study aims to investigate cognitive multitasking performance as a function of (1) virtual reality and computer administration and (2) gender differences. To explore this issue, subjects were asked to perform simultaneous tasks (span forward and backward, logical–arithmetic reasoning, and visuospatial reasoning) in virtual reality via a head-mounted display system (HDMS) and on a personal computer (PC). Our results showed in virtual reality an overall impairment of executive functioning but a better performance of women, compared to men, in visuospatial reasoning. These findings are consistent with previous studies showing a detrimental effect of virtual reality on cognitive functioning. Full article

(This article belongs to the Special Issue Selected Papers from 1st International Workshop on Affective Computing and Health Care: New Research and Industrial Perspectives)

► Show Figures

Figure 1

12 pages, 2446 KiB

Open AccessArticle

Short-Term l-arginine Treatment Mitigates Early Damage of Dermal Collagen Induced by Diabetes

by Irena Miler, Mihailo D. Rabasovic, Sonja Askrabic, Andreas Stylianou, Bato Korac and Aleksandra Korac

Bioengineering 2024, 11(4), 407; https://doi.org/10.3390/bioengineering11040407 - 21 Apr 2024

Viewed by 1437

Abstract

Changes in the structural properties of the skin due to collagen alterations are an important factor in diabetic skin complications. Using a combination of photonic methods as an optic diagnostic tool, we investigated the structural alteration in rat dermal collagen I in diabetes, [...] Read more.

Changes in the structural properties of the skin due to collagen alterations are an important factor in diabetic skin complications. Using a combination of photonic methods as an optic diagnostic tool, we investigated the structural alteration in rat dermal collagen I in diabetes, and after short-term l-arginine treatment. The multiplex approach shows that in the early phase of diabetes, collagen fibers are partially damaged, resulting in the heterogeneity of fibers, e.g., “patchy patterns” of highly ordered/disordered fibers, while l-arginine treatment counteracts to some extent the conformational changes in collagen-induced by diabetes and mitigates the damage. Raman spectroscopy shows intense collagen conformational changes via amides I and II in diabetes, suggesting that diabetes-induced structural changes in collagen originate predominantly from individual collagen molecules rather than supramolecular structures. There is a clear increase in the amounts of newly synthesized proline and hydroxyproline after treatment with l-arginine, reflecting the changed collagen content. This suggests that it might be useful for treating and stopping collagen damage early on in diabetic skin. Our results demonstrate that l-arginine attenuates the early collagen I alteration caused by diabetes and that it could be used to treat and prevent collagen damage in diabetic skin at a very early stage. Full article

(This article belongs to the Section Biomedical Engineering and Biomaterials)

► Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
pSHG images of dermal collagen (a) and calculated β values (b) of control (non-diabetic), l-arginine (l-arginine-treated non-diabetic), diabetic, and diabetic l-arginine-treated rats. Different colors in panel (a) indicate the different values of the β-coefficient (as indicated by the color bar in the leftmost panel). Thus, greener images indicate a higher value of the β-coefficient and, therefore, more arranged collagen. The opposite is true for red images. The values represent the mean ± SEM, for each point n = 27, * p < 0.05. Scale bars: 50 µm. Full article ">Figure 2
Mean Raman spectra of dermal collagen I structure of control (non-diabetic), l-arginine (l-arginine-treated non-diabetic), diabetic, and diabetic l-arginine-treated rats. The light gray highlighted areas are magnified and present the most intensive changes described in <a href="#app1-bioengineering-11-00407" class="html-app">Table S1</a>. Full article ">Figure 3
Raman spectra of dermal collagen I of control (non-diabetic), l-arginine (l-arginine−treated non-diabetic), diabetic, and diabetic l-arginine-treated rats. Full article ">Figure 4
Young’s modulus measured by AFM in control (non-diabetic), l-arginine (l-arginine-treated non-diabetic), diabetic, and diabetic l-arginine-treated rats. The values represent the mean ± SEM, * p < 0.05. Full article ">Figure 5
Micrographs of the rat skin of control (non-diabetic), l-arginine (l-arginine-treated non-diabetic), diabetic, and diabetic l-arginine-treated rats stained with picrosirius red and analyzed without polarizers (left images) and with linear polarizers (right images). Magnification: 10×; original scale bars: 250 µm. Full article ">

21 pages, 3240 KiB

Open AccessArticle

Deep Transfer Learning Using Real-World Image Features for Medical Image Classification, with a Case Study on Pneumonia X-ray Images

by Chanhoe Gu and Minhyeok Lee

Bioengineering 2024, 11(4), 406; https://doi.org/10.3390/bioengineering11040406 - 20 Apr 2024

Cited by 1 | Viewed by 1620

Abstract

Deep learning has profoundly influenced various domains, particularly medical image analysis. Traditional transfer learning approaches in this field rely on models pretrained on domain-specific medical datasets, which limits their generalizability and accessibility. In this study, we propose a novel framework called real-world feature [...] Read more.

Deep learning has profoundly influenced various domains, particularly medical image analysis. Traditional transfer learning approaches in this field rely on models pretrained on domain-specific medical datasets, which limits their generalizability and accessibility. In this study, we propose a novel framework called real-world feature transfer learning, which utilizes backbone models initially trained on large-scale general-purpose datasets such as ImageNet. We evaluate the effectiveness and robustness of this approach compared to models trained from scratch, focusing on the task of classifying pneumonia in X-ray images. Our experiments, which included converting grayscale images to RGB format, demonstrate that real-world-feature transfer learning consistently outperforms conventional training approaches across various performance metrics. This advancement has the potential to accelerate deep learning applications in medical imaging by leveraging the rich feature representations learned from general-purpose pretrained models. The proposed methodology overcomes the limitations of domain-specific pretrained models, thereby enabling accelerated innovation in medical diagnostics and healthcare. From a mathematical perspective, we formalize the concept of real-world feature transfer learning and provide a rigorous mathematical formulation of the problem. Our experimental results provide empirical evidence supporting the effectiveness of this approach, laying the foundation for further theoretical analysis and exploration. This work contributes to the broader understanding of feature transferability across domains and has significant implications for the development of accurate and efficient models for medical image analysis, even in resource-constrained settings. Full article

(This article belongs to the Special Issue Artificial Intelligence in Advanced Medical Imaging - 2nd Edition)

► Show Figures

Figure 1

11 pages, 4904 KiB

Open AccessArticle

Breaking the Limit of Cardiovascular Regenerative Medicine: Successful 6-Month Goat Implant in World’s First Ascending Aortic Replacement Using Biotube Blood Vessels

by Kazuki Mori, Tadashi Umeno, Takayuki Kawashima, Tomoyuki Wada, Takuro Genda, Masanagi Arakura, Yoshifumi Oda, Takayuki Mizoguchi, Ryosuke Iwai, Tsutomu Tajikawa, Yasuhide Nakayama and Shinji Miyamoto

Bioengineering 2024, 11(4), 405; https://doi.org/10.3390/bioengineering11040405 - 20 Apr 2024

Cited by 2 | Viewed by 1160

Abstract

This study investigated six-month outcomes of first models of ascending aortic replacement. The molds used to produce the Biotube were implanted subcutaneously in goats. After 2–3 months, the molds were explanted to obtain the Biotubes (inner diameter, 12 mm; wall thickness, 1.5 mm). [...] Read more.

This study investigated six-month outcomes of first models of ascending aortic replacement. The molds used to produce the Biotube were implanted subcutaneously in goats. After 2–3 months, the molds were explanted to obtain the Biotubes (inner diameter, 12 mm; wall thickness, 1.5 mm). Next, we performed ascending aortic replacement using the Biotube in five allogenic goats. At 6 months, the animals underwent computed tomography (CT) and histologic evaluation. As a comparison, we performed similar surgeries using glutaraldehyde-fixed autologous pericardial rolls or pig-derived heterogenous Biotubes. At 6 months, CT revealed no aneurysmalization of the Biotube or pseudoaneurysm formation. The histologic evaluation showed development of endothelial cells, smooth muscle cells, and elastic fibers along the Biotube. In the autologous pericardium group, there was no evidence of new cell development, but there was calcification. The histologic changes observed in the heterologous Biotube group were similar to those in the allogenic Biotube group. However, there was inflammatory cell infiltration in some heterologous Biotubes. Based on the above, we could successfully create the world’s first Biotube-based ascending aortic replacement models. The present results indicate that the Biotube may serve as a scaffold for aortic tissue regeneration. Full article

(This article belongs to the Section Regenerative Engineering)

► Show Figures

Figure 1

10 pages, 1864 KiB

Open AccessArticle

An Investigation of Running Kinematics with Recovered Anterior Cruciate Ligament Reconstruction on a Treadmill and In-Field Using Inertial Measurement Units: A Preliminary Study

by Matteo Hill, Pierre Kiesewetter, Thomas L. Milani and Christian Mitschke

Bioengineering 2024, 11(4), 404; https://doi.org/10.3390/bioengineering11040404 - 19 Apr 2024

Viewed by 1258

Abstract

Anterior cruciate ligament reconstruction (ACLR) may affect movement even years after surgery. The purpose of this study was to determine possible interlimb asymmetries due to ACLR when running on a treadmill and in field conditions, with the aim of contributing to the establishment [...] Read more.

Anterior cruciate ligament reconstruction (ACLR) may affect movement even years after surgery. The purpose of this study was to determine possible interlimb asymmetries due to ACLR when running on a treadmill and in field conditions, with the aim of contributing to the establishment of objective movement assessment in real-world settings; moreover, we aimed to gain knowledge on recovered ACLR as a biomechanical risk factor. Eight subjects with a history of unilateral ACLR 5.4 ± 2.8 years after surgery and eight healthy subjects ran 1 km on a treadmill and 1 km on a concrete track. The ground contact time and triaxial peak tibial accelerations were recorded using inertial measurement units. Interlimb differences within subjects were tested and compared between conditions. There were no significant differences between limbs in the ACLR subjects or in healthy runners for any of the chosen parameters on both running surfaces. However, peak tibial accelerations were higher during field running (p-values < 0.01; Cohen’s d effect sizes > 0.8), independent of health status. To minimize limb loading due to higher impacts during field running, this should be considered when choosing a running surface, especially in rehabilitation or when running with a minor injury or health issues. Full article

(This article belongs to the Special Issue Biomechanics of Human Movement and Its Clinical Applications)

► Show Figures

Figure 1

17 pages, 7407 KiB

Open AccessCase Report

Biomechanical Considerations in the Orthodontic Treatment of a Patient with Stabilised Stage IV Grade C Generalised Periodontitis: A Case Report

by Fung Hou Kumoi Mineaki Howard Sum, Zhiyi Shan, Yat Him Dave Chan, Ryan Julian Dick Hei Chu, George Pelekos and Tsang Tsang She

Bioengineering 2024, 11(4), 403; https://doi.org/10.3390/bioengineering11040403 - 19 Apr 2024

Viewed by 1375

Abstract

Orthodontic treatment of periodontally compromised patients presents unique challenges, including controlling periodontal inflammation, applying appropriate force, designing an effective dental anchorage, and maintaining treatment results. Deteriorated periodontal support leads to alterations in the biological responses of teeth to mechanical forces, and thus orthodontists [...] Read more.

Orthodontic treatment of periodontally compromised patients presents unique challenges, including controlling periodontal inflammation, applying appropriate force, designing an effective dental anchorage, and maintaining treatment results. Deteriorated periodontal support leads to alterations in the biological responses of teeth to mechanical forces, and thus orthodontists must take greater care when treating patients with periodontal conditions than when treating those with a good periodontal status. In this article, we report the case of a 59-year-old woman with stabilised Stage IV grade C generalised periodontitis characterised by pathological tooth migration (PTM). The assessment, planning, and treatment of this patient with orthodontic fixed appliances is described. Moreover, the anchorage planning and biomechanical considerations are detailed. Specific orthodontic appliances were employed in this case to produce force systems for achieving precise tooth movement, which included a cantilever, mini-screws, and a box loop. Careful application of those appliances resulted in satisfactory aesthetic and functional orthodontic outcomes in the patient. This case highlights the importance of multidisciplinary collaboration in the treatment of patients with severe periodontitis and the potential for tailored biomechanical approaches in orthodontic treatment to furnish good outcomes. Full article

(This article belongs to the Special Issue Application of Bioengineering to Clinical Orthodontics)

► Show Figures

Figure 1

19 pages, 10631 KiB

Open AccessArticle

Dimensionality Matters: Exploiting UV-Photopatterned 2D and Two-Photon-Printed 2.5D Contact Guidance Cues to Control Corneal Fibroblast Behavior and Collagen Deposition

by Cas van der Putten, Gozde Sahin, Rhiannon Grant, Mirko D’Urso, Stefan Giselbrecht, Carlijn V. C. Bouten and Nicholas A. Kurniawan

Bioengineering 2024, 11(4), 402; https://doi.org/10.3390/bioengineering11040402 - 19 Apr 2024

Cited by 1 | Viewed by 1784

Abstract

In the event of disease or injury, restoration of the native organization of cells and extracellular matrix is crucial for regaining tissue functionality. In the cornea, a highly organized collagenous tissue, keratocytes can align along the anisotropy of the physical microenvironment, providing a [...] Read more.

In the event of disease or injury, restoration of the native organization of cells and extracellular matrix is crucial for regaining tissue functionality. In the cornea, a highly organized collagenous tissue, keratocytes can align along the anisotropy of the physical microenvironment, providing a blueprint for guiding the organization of the collagenous matrix. Inspired by this physiological process, anisotropic contact guidance cues have been employed to steer the alignment of keratocytes as a first step to engineer in vitro cornea-like tissues. Despite promising results, two major hurdles must still be overcome to advance the field. First, there is an enormous design space to be explored in optimizing cellular contact guidance in three dimensions. Second, the role of contact guidance cues in directing the long-term deposition and organization of extracellular matrix proteins remains unknown. To address these challenges, here we combined two microengineering strategies—UV-based protein patterning (2D) and two-photon polymerization of topographies (2.5D)—to create a library of anisotropic contact guidance cues with systematically varying height (H, 0 µm ≤ H ≤ 20 µm) and width (W, 5 µm ≤ W ≤ 100 µm). With this unique approach, we found that, in the short term (24 h), the orientation and morphology of primary human fibroblastic keratocytes were critically determined not only by the pattern width, but also by the height of the contact guidance cues. Upon extended 7-day cultures, keratocytes were shown to produce a dense, fibrous collagen network along the direction of the contact guidance cues. Moreover, increasing the heights also increased the aligned fraction of deposited collagen and the contact guidance response of cells, all whilst the cells maintained the fibroblastic keratocyte phenotype. Our study thus reveals the importance of dimensionality of the physical microenvironment in steering both cellular organization and the formation of aligned, collagenous tissues. Full article

(This article belongs to the Special Issue Bioengineering and the Eye—2nd Edition)

► Show Figures

Graphical abstract

16 pages, 1642 KiB

Open AccessArticle

Assessment of Microvascular Hemodynamic Adaptations in Finger Flexors of Climbers

by Blai Ferrer-Uris, Albert Busquets, Faruk Beslija and Turgut Durduran

Bioengineering 2024, 11(4), 401; https://doi.org/10.3390/bioengineering11040401 - 19 Apr 2024

Viewed by 1275

Abstract

Climbing performance is greatly dependent on the endurance of the finger flexors which, in turn, depends on the ability to deliver and use oxygen within the muscle. Near-infrared spectroscopy (NIRS) and diffuse correlation spectroscopy (DCS) have provided new possibilities to explore these phenomena [...] Read more.

Climbing performance is greatly dependent on the endurance of the finger flexors which, in turn, depends on the ability to deliver and use oxygen within the muscle. Near-infrared spectroscopy (NIRS) and diffuse correlation spectroscopy (DCS) have provided new possibilities to explore these phenomena in the microvascular environment. The aim of the present study was to explore climbing-related microvascular adaptations through the comparison of the oxygen concentration and hemodynamics of the forearm between climbers and non-climber active individuals during a vascular occlusion test (VOT). Seventeen climbers and fifteen non-climbers joined the study. Through NIRS and DCS, the oxyhemoglobin (O₂Hb) and deoxyhemoglobin (HHb) concentrations, tissue saturation index (TSI), and blood flow index (BFI) were obtained from the flexor digitorum profundus during the VOT. During the reactive hyperemia, climbers presented greater blood flow slopes (p = 0.043, d = 0.573), as well as greater O₂Hb maximum values (p = 0.001, d = 1.263) and HHb minimum values (p = 0.009, d = 0.998), than non-climbers. The superior hemodynamics presented by climbers could indicate potential training-induced structural and functional adaptations that could enhance oxygen transportation to the muscle, and thus enhance muscle endurance and climbing performance. Full article

(This article belongs to the Special Issue Biomechanics and Sports Medicine in Sport Climbing, Featuring Selected Contributions Presented at the 6th IRCRA Congress 2023)

► Show Figures

Figure 1

Figure 1
Example from a single participant of the (A) oxyhemoglobin concentration (O2Hb), (B) deoxyhemoglobin concentration (HHb), (C) tissue saturation index (TSI), and (D) blood flow index (BFI) changes during the vascular occlusion test (VOT). Vertical lines indicate the inflation and release of the pressure cuff, which were used to divide the test into the baseline phase (BA), occlusion phase (OC), and reactive hyperemia phase (HY). Calculated variables for each signal are schematically represented. Abbreviations: rBF: relative blood flow, tmin: time to minimum value, tmax: time to maximum value, HTR: half-time to recovery. Full article ">Figure 2
Normalized oxyhemoglobin (∆O2Hb, (A)), deoxyhemoglobin (∆HHb, (B)), tissue saturation index (∆TSI, (C)), and relative blood flow (rBF, (D)) example traces from a climber (left side) and a non-climber (right side) participant. Full article ">Figure 3
Group means and standard deviations, and individual subject’s values from the maximum delta hemoglobin concentration (HY-∆O2Hbmax, (A)), the minimum delta deoxyhemoglobin concentration (HY-∆HHbmin, (B)), and the relative blood flow slope (HY-rBFslope, (C)) during the hyperemia phase (HY) of the vascular occlusion test. Significant group differences were found for the three parameters (p < 0.05). When DCS-related variables during the HY phase were explored, no main group effect was found (F (1, 30) = 2.046, p = 0.130, η2p = 0.180, power = 0.468), indicating that groups presented similar blood flow dynamics when compared using the compound of DCS variables. Nevertheless, when pairwise comparisons were explored, climbers showed a greater HY-rBFslope compared to non-climbers (p = 0.043, d = 0.573), indicating a faster increase in blood flow after the occlusive pressure release (<a href="#bioengineering-11-00401-f003" class="html-fig">Figure 3</a> and as can been seen in <a href="#bioengineering-11-00401-f002" class="html-fig">Figure 2</a> with sample data from each group). Despite this difference, similar HY-rBFmax, HY-rBFtmax, and HY-rBFHTR were found between groups. Full article ">Figure 4
Schematic representation summarizing our study methodological approach, main findings, and hypotheses. Full article ">

21 pages, 1468 KiB

Open AccessArticle

Ergonomic Analysis of Dental Work in Different Oral Quadrants: A Motion Capture Preliminary Study among Endodontists

by Sophie Feige, Fabian Holzgreve, Laura Fraeulin, Christian Maurer-Grubinger, Werner Betz, Christina Erbe, Albert Nienhaus, David A. Groneberg and Daniela Ohlendorf

Bioengineering 2024, 11(4), 400; https://doi.org/10.3390/bioengineering11040400 - 19 Apr 2024

Cited by 1 | Viewed by 1530

Abstract

Background: Dentists, including endodontists, frequently experience musculoskeletal disorders due to unfavourable working postures. Several measures are known to reduce the ergonomic risk; however, there are still gaps in the research, particularly in relation to dental work in the different oral regions (Quadrants 1–4). [...] Read more.

Background: Dentists, including endodontists, frequently experience musculoskeletal disorders due to unfavourable working postures. Several measures are known to reduce the ergonomic risk; however, there are still gaps in the research, particularly in relation to dental work in the different oral regions (Quadrants 1–4). Methods: In this study (of a pilot character), a total of 15 dentists (8 male and 7 female) specialising in endodontics were measured while performing root canal treatments on a phantom head. These measurements took place in a laboratory setting using an inertial motion capture system. A slightly modified Rapid Upper Limb Assessment (RULA) coding system was employed for the analysis of kinematic data. The significance level was set at p = 0.05. Results: The ergonomic risk for the entire body was higher in the fourth quadrant than in the first quadrant for 80% of the endodontists and higher than in the second quadrant for 87%. For 87% of the endodontists, the ergonomic risk for the right side of the body was significantly higher in the fourth quadrant compared to the first and second quadrant. The right arm was stressed more in the lower jaw than in the upper jaw, and the neck also showed a greater ergonomic risk in the fourth quadrant compared to the first quadrant. Conclusion: In summary, both the total RULA score and scores for the right- and lefthand sides of the body ranged between 5 and 6 out of a possible 7 points. Considering this considerable burden, heightened attention, especially to the fourth quadrant with a significantly higher ergonomic risk compared to Quadrants 1 and 2, may be warranted. Full article

(This article belongs to the Special Issue Advances in Wearable Technologies for ”In-Field” Assessment of Biomechanical Risk)

► Show Figures

Figure 1

19 pages, 5200 KiB

Open AccessArticle

Precision Identification of Locally Advanced Rectal Cancer in Denoised CT Scans Using EfficientNet and Voting System Algorithms

by Chun-Yu Lin, Jacky Chung-Hao Wu, Yen-Ming Kuan, Yi-Chun Liu, Pi-Yi Chang, Jun-Peng Chen, Henry Horng-Shing Lu and Oscar Kuang-Sheng Lee

Bioengineering 2024, 11(4), 399; https://doi.org/10.3390/bioengineering11040399 - 19 Apr 2024

Cited by 1 | Viewed by 1515

Abstract

Background and objective: Local advanced rectal cancer (LARC) poses significant treatment challenges due to its location and high recurrence rates. Accurate early detection is vital for treatment planning. With magnetic resonance imaging (MRI) being resource-intensive, this study explores using artificial intelligence (AI) to [...] Read more.

Background and objective: Local advanced rectal cancer (LARC) poses significant treatment challenges due to its location and high recurrence rates. Accurate early detection is vital for treatment planning. With magnetic resonance imaging (MRI) being resource-intensive, this study explores using artificial intelligence (AI) to interpret computed tomography (CT) scans as an alternative, providing a quicker, more accessible diagnostic tool for LARC. Methods: In this retrospective study, CT images of 1070 T3–4 rectal cancer patients from 2010 to 2022 were analyzed. AI models, trained on 739 cases, were validated using two test sets of 134 and 197 cases. By utilizing techniques such as nonlocal mean filtering, dynamic histogram equalization, and the EfficientNetB0 algorithm, we identified images featuring characteristics of a positive circumferential resection margin (CRM) for the diagnosis of locally advanced rectal cancer (LARC). Importantly, this study employs an innovative approach by using both hard and soft voting systems in the second stage to ascertain the LARC status of cases, thus emphasizing the novelty of the soft voting system for improved case identification accuracy. The local recurrence rates and overall survival of the cases predicted by our model were assessed to underscore its clinical value. Results: The AI model exhibited high accuracy in identifying CRM-positive images, achieving an area under the curve (AUC) of 0.89 in the first test set and 0.86 in the second. In a patient-based analysis, the model reached AUCs of 0.84 and 0.79 using a hard voting system. Employing a soft voting system, the model attained AUCs of 0.93 and 0.88, respectively. Notably, AI-identified LARC cases exhibited a significantly higher five-year local recurrence rate and displayed a trend towards increased mortality across various thresholds. Furthermore, the model’s capability to predict adverse clinical outcomes was superior to those of traditional assessments. Conclusion: AI can precisely identify CRM-positive LARC cases from CT images, signaling an increased local recurrence and mortality rate. Our study presents a swifter and more reliable method for detecting LARC compared to traditional CT or MRI techniques. Full article

(This article belongs to the Special Issue Application of Deep Learning in Medical Diagnosis)

► Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
Diagrammatic representation of study material workflow. This figure outlines the systematic process used in this study, tracing the flow of materials from initial data collection through to their final application within the research framework. Full article ">Figure 2
Characterization and processing of rectal cancer features in CT imaging. (a) Criteria for identifying CRM-threatening features associated with locally advanced rectal cancer in CT scans; (b) illustrative cases of rectal cancer in CT imagery lacking CRM-threatening attributes; (c) comparative illustration of three cropping techniques for rectal cancer image preparation: full image, external pelvic boundary, and internal pelvic boundary. Full article ">Figure 3
AI-based diagnostic framework for detecting the locally advanced rectal cancer architectural blueprint of the AI model for LARC identification. This diagram illustrates the methodological approach and the deep learning architecture utilized in this study for the detection of locally advanced rectal cancer (LARC) from CT scans. The framework encapsulates the AI’s training and predictive process, detailing the progression from image preprocessing to the final LARC classification. Full article ">Figure 4
Optimal diagnostic outcomes using Models 2a and 2b. The green dotted line indicate randomize classifier. (a–c) Performance of Model 2a, amalgamating data from the training set and test set 2, when applied to test set 1. (a) Identification of CRM-threatening features indicative of LARC; (b) determination of LARC status using a hard voting threshold of one-third; (c) assessment of LARC cases employing a soft voting threshold of a half. (d–f) Efficacy of Model 2b, integrating data from the training set and test set 1, utilized on test set 2. (d) Image analysis for CRM-threatening features associated with LARC; (e) LARC case adjudication based on a hard voting threshold of one-fourth; (f) LARC case determination via a soft voting threshold of one-third. Full article ">Figure 5
Five-year local recurrence and overall survival rates for LARC from physician and analysis of different AI methodologies. Full article ">Figure 6
Three visual examples of interpretation results by a physician and AI. The surgery outcome with a positive pathological circumferential margin (pCRM), disease survival of local recurrence, and overall survival time were also recorded. The following abbreviations are used: pCRM+, positive pathological circumferential margin; LR, local recurrence; OS, overall survival. Full article ">

13 pages, 1278 KiB

Open AccessArticle

Effects of Action Observation Plus Motor Imagery Administered by Immersive Virtual Reality on Hand Dexterity in Healthy Subjects

by Paola Adamo, Gianluca Longhi, Federico Temporiti, Giorgia Marino, Emilia Scalona, Maddalena Fabbri-Destro, Pietro Avanzini and Roberto Gatti

Bioengineering 2024, 11(4), 398; https://doi.org/10.3390/bioengineering11040398 - 19 Apr 2024

Viewed by 1188

Abstract

Action observation and motor imagery (AOMI) are commonly delivered through a laptop screen. Immersive virtual reality (VR) may enhance the observer’s embodiment, a factor that may boost AOMI effects. The study aimed to investigate the effects on manual dexterity of AOMI delivered through [...] Read more.

Action observation and motor imagery (AOMI) are commonly delivered through a laptop screen. Immersive virtual reality (VR) may enhance the observer’s embodiment, a factor that may boost AOMI effects. The study aimed to investigate the effects on manual dexterity of AOMI delivered through immersive VR compared to AOMI administered through a laptop. To evaluate whether VR can enhance the effects of AOMI, forty-five young volunteers were enrolled and randomly assigned to the VR-AOMI group, who underwent AOMI through immersive VR, the AOMI group, who underwent AOMI through a laptop screen, or the control group, who observed landscape video clips. All participants underwent a 5-day treatment, consisting of 12 min per day. We investigated between and within-group differences after treatments relative to functional manual dexterity tasks using the Purdue Pegboard Test (PPT). This test included right hand (R), left hand (L), both hands (B), R + L + B, and assembly tasks. Additionally, we analyzed kinematics parameters including total and sub-phase duration, peak and mean velocity, and normalized jerk, during the Nine-Hole Peg Test to examine whether changes in functional scores may also occur through specific kinematic patterns. Participants were assessed at baseline (T0), after the first training session (T1), and at the end of training (T2). A significant time by group interaction and time effects were found for PPT, where both VR-AOMI and AOMI groups improved at the end of training. Larger PPT-L task improvements were found in the VR-AOMI group (d: 0.84, CI₉₅: 0.09–1.58) compared to the AOMI group from T0 to T1. Immersive VR used for the delivery of AOMI speeded up hand dexterity improvements. Full article

(This article belongs to the Special Issue Bioengineering of the Motor System)

► Show Figures

Graphical abstract

12 pages, 1521 KiB

Open AccessArticle

Micro-Computed Tomography Analysis of Peri-Implant Bone Defects Exposed to a Peri-Implantitis Microcosm, with and without Bone Substitute, in a Rabbit Model: A Pilot Study

by Camila Panes, Iván Valdivia-Gandur, Carlos Veuthey, Vanessa Sousa, Mariano del Sol and Víctor Beltrán

Bioengineering 2024, 11(4), 397; https://doi.org/10.3390/bioengineering11040397 - 19 Apr 2024

Viewed by 1288

Abstract

Peri-implantitis is an inflammatory condition characterized by inflammation in the peri-implant connective tissue and a progressive loss of supporting bone; it is commonly associated with the presence of biofilms on the surface of the implant, which is an important factor in the development [...] Read more.

Peri-implantitis is an inflammatory condition characterized by inflammation in the peri-implant connective tissue and a progressive loss of supporting bone; it is commonly associated with the presence of biofilms on the surface of the implant, which is an important factor in the development and progression of the disease. The objective of this study was to evaluate, using micro-CT, the bone regeneration of surgically created peri-implant defects exposed to a microcosm of peri-implantitis. Twenty-three adult New Zealand white rabbits were included in the study. Bone defects of 7 mm diameter were created in both tibiae, and a cap-shaped titanium device was placed in the center, counter-implanted with a peri-implantitis microcosm. The bone defects received a bone substitute and/or a resorbable synthetic PLGA membrane, according to random distribution. Euthanasia was performed 15 and 30 days postoperatively. Micro-CT was performed on all samples to quantify bone regeneration parameters. Bone regeneration of critical defects occurred in all experimental groups, with a significantly greater increase in cases that received bone graft treatment (p < 0.0001), in all measured parameters, at 15 and 30 days. No significant differences were observed in the different bone neoformation parameters between the groups that did not receive bone grafts (p > 0.05). In this experimental model, the presence of peri-implantitis microcosms was not a determining factor in the bone volume parameter, both in the groups that received regenerative treatment and in those that did not. Full article

(This article belongs to the Special Issue Biomaterials for Bone Repair and Regeneration)

► Show Figures

Figure 1

Figure 1
Schematization of the division of the experimental groups. Full article ">Figure 2
Example of the spatial reorientation applied to each scanned sample using DataViewer v.1.5.6.2 software (Bruker-microCT, Kontich, Bélgica), matching the axial axis and center of the implant. The images oriented in the coronal plane were extracted for subsequent segmentation and analysis of the images. COR: coronal plane; TRA: transverse plane; SAG: sagittal plane. Full article ">Figure 3
Graphs of the parameters obtained by computerized microtomography, 15 days post-surgery: (A) Bone volume (BV) present in the VOI. (B) Percent bone volume (BV/TV). (C) Bone surface (BS). (D) Bone surface/volume ratio (BS/BV). (E) Bone surface density (BS/TV). ** p ≤ 0.001; *** p ≤ 0.0001. PLGA: PLGA membrane; MB: membrane and bone graft; BG: bone graft; CRT: control group; UC: untreated cap. Full article ">Figure 4
Graphs of the parameters obtained by computerized microtomography, 30 days post-surgery: (A) Bone volume (BV) present in the VOI. (B) Percent bone volume (BV/TV). (C) Bone surface (BS). (D) Bone surface/volume ratio (BS/BV). (E) Bone surface density (BS/TV). * p ≤ 0.01; ** p ≤ 0.001; *** p ≤ 0.0001. PLGA: PLGA membrane; MB: membrane and bone graft; BG: bone graft; CRT: control group; UC: untreated cap. Full article ">

12 pages, 738 KiB

Open AccessReview

A Systematic Review of Bone Bruise Patterns following Acute Anterior Cruciate Ligament Tears: Insights into the Mechanism of Injury

by Sueen Sohn, Saad Mohammed AlShammari, Byung Jun Hwang and Man Soo Kim

Bioengineering 2024, 11(4), 396; https://doi.org/10.3390/bioengineering11040396 - 19 Apr 2024

Viewed by 1182

Abstract

(1) Background: The purpose of this systematic review was to determine the prevalence of bone bruises in patients with anterior cruciate ligament (ACL) injuries and the location of the bruises relative to the tibia and femur. Understanding the relative positions of these bone [...] Read more.

(1) Background: The purpose of this systematic review was to determine the prevalence of bone bruises in patients with anterior cruciate ligament (ACL) injuries and the location of the bruises relative to the tibia and femur. Understanding the relative positions of these bone bruises could enhance our comprehension of the knee loading patterns that occur during an ACL injury. (2) Methods: The MEDLINE, EMBASE, and the Cochrane Library databases were searched for studies that evaluated the presence of bone bruises following ACL injuries. Study selection, data extraction, and a systematic review were performed. (3) Results: Bone bruises were observed in 3207 cases (82.8%) at the lateral tibia plateau (LTP), 1608 cases (41.5%) at the medial tibia plateau (MTP), 2765 cases (71.4%) at the lateral femoral condyle (LFC), and 1257 cases (32.4%) at the medial femoral condyle (MFC). Of the 30 studies, 11 were able to assess the anterior to posterior direction. The posterior LTP and center LFC were the most common areas of bone bruises. Among the 30 studies, 14 documented bone bruises across all four sites (LTP, MTP, LFC, and MFC). The most common pattern was bone bruises appearing at the LTP and LFC. (4) Conclusions: The most frequently observed pattern of bone bruises was restricted to the lateral aspects of both the tibia and femur. In cases where bone bruises were present on both the lateral and medial sides, those on the lateral side exhibited greater severity. The positioning of bone bruises along the front–back axis indicated a forward shift of the tibia in relation to the femur during ACL injuries. Full article

(This article belongs to the Special Issue Biomechanics of Sports Injuries)

► Show Figures

Figure 1

12 pages, 2260 KiB

Open AccessArticle

Tensile Yield Strain of Human Cortical Bone from the Femoral Diaphysis Is Constant among Healthy Adults and across the Anatomical Quadrants

by Massimiliano Baleani, Paolo Erani, Alice Acciaioli and Enrico Schileo

Bioengineering 2024, 11(4), 395; https://doi.org/10.3390/bioengineering11040395 - 19 Apr 2024

Viewed by 1242

Abstract

The literature suggests that the yield strain of cortical bone is invariant to its stiffness (elastic modulus) and strength (yield stress). However, data about intra-individual variations, e.g., the influence of different collagen/mineral organisations observed in bone aspects withstanding different habitual loads, are lacking. [...] Read more.

The literature suggests that the yield strain of cortical bone is invariant to its stiffness (elastic modulus) and strength (yield stress). However, data about intra-individual variations, e.g., the influence of different collagen/mineral organisations observed in bone aspects withstanding different habitual loads, are lacking. The hypothesis that the yield strain of human cortical bone tissue, retrieved from femoral diaphyseal quadrants subjected to different habitual loads, is invariant was tested. Four flat dumbbell-shaped specimens were machined from each quadrant of the proximal femoral diaphysis of five adult donors for a total of 80 specimens. Two extensometers attached to the narrow specimen region were used to measure deformation during monotonic tensile testing. The elastic modulus (linear part of the stress–strain curve) and yield strain/stress at a 0.2% offset were obtained. Elastic modulus and yield stress values were, respectively, in the range of 12.2–20.5 GPa and 75.9–136.6 MPa and exhibited a positive linear correlation. All yield strain values were in the narrow range of 0.77–0.87%, regardless of the stiffness and strength of the tissue and the anatomical quadrant. In summary, the results corroborate the hypothesis that tensile yield strain in cortical bone is invariant, irrespective also of the anatomical quadrant. The mean yield strain value found in this study is similar to what was reported by inter-species and evolution studies but slightly higher than previous reports in humans, possibly because of the younger age of our subjects. Further investigations are needed to elucidate a possible dependence of yield strain on age. Full article

(This article belongs to the Section Biomechanics and Sports Medicine)

► Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
(I) A 46 mm thick slice is cut from the proximal femoral diaphysis. (II) The distal end of the diaphyseal slice is embedded in acrylic resin. (III) Six cuts are made in the proximal–distal direction, parallel to the frontal plane of the femur. (IV) Six cuts are made in the proximal–distal direction, parallel to the sagittal plane of the femur. (V) A cut is made orthogonal to the axis of the slice at 35 mm from the proximal end. (VI) Sketch showing the dimensions of the cortical slices that were obtained. (VII) The contour of the narrow part of the sample is machined with a diamond-coated tool. The periosteal-side contour of the narrow section was machined 0.7 mm inside the original periosteal surface. (VIII) The surface of the flat dumbbell-shaped specimen is polished by removing 0.05 mm of cortical bone tissue. (IX) Sketch showing the dimensions of the narrow part of flat dumbbell-shaped specimens that were obtained (note: because of the local anatomy of the cortical wall, non-standard (i.e., smaller with a 10:1 ratio of gauge length to thickness and 2.5:1 ratio of gauge length to width) dumbbell specimens were obtained). Full article ">Figure 2
(a) The diaphyseal slice with six cuts in the proximal–distal direction, parallel to the frontal plane of the femur (step III in <a href="#bioengineering-11-00395-f001" class="html-fig">Figure 1</a>). (b) A cortical slice retrieved from the cortical wall (step VI in <a href="#bioengineering-11-00395-f001" class="html-fig">Figure 1</a>). No trabecular bone tissue is present on the endosteal side. (c) The flat dumbbell-shaped specimen obtained at the end of the procedure (step IX in <a href="#bioengineering-11-00395-f001" class="html-fig">Figure 1</a>). The periosteal side (left edge of the specimen) of the gripping part was not machined. Therefore, irregularities in the left contour of the gripping part of the specimen are visible. Full article ">Figure 3
Boxplot of yield strain values (%) by quadrant. The central line in each box indicates the median value, the boxes represent the 25th and 75th percentile, and the whiskers represent the 10th and 90th percentiles. Full article ">Figure 4
Boxplot of yield stress values (MPa) by quadrant. The central line in each box indicates the median value, the boxes represent the 25th and 75th percentile, and the whiskers represent the 10th and 90th percentiles. Full article ">Figure 5
Boxplot of elastic modulus values (GPa) by quadrant. The central line in each box indicates the median value, the boxes represent the 25th and 75th percentile, and the whiskers represent the 10th and 90th percentiles. Full article ">Figure 6
Scatterplot of yield strain versus elastic modulus for all 80 specimens. The line represents the correlation mentioned in the text. Full article ">Figure 7
Scatterplot of yield stress versus elastic modulus for all 80 specimens. The line represents the correlation mentioned in the text. Full article ">

12 pages, 6255 KiB

Open AccessArticle

Finite Element Analysis of Fixed Orthodontic Retainers

by Sebastian Hetzler, Stefan Rues, Andreas Zenthöfer, Peter Rammelsberg, Christopher J. Lux and Christoph J. Roser

Bioengineering 2024, 11(4), 394; https://doi.org/10.3390/bioengineering11040394 - 18 Apr 2024

Viewed by 1425

Abstract

The efficacy of retainers is a pivotal concern in orthodontic care. This study examined the biomechanical behaviour of retainers, particularly the influence of retainer stiffness and tooth resilience on force transmission and stress distribution. To do this, a finite element model was created [...] Read more.

The efficacy of retainers is a pivotal concern in orthodontic care. This study examined the biomechanical behaviour of retainers, particularly the influence of retainer stiffness and tooth resilience on force transmission and stress distribution. To do this, a finite element model was created of the lower jaw from the left to the right canine with a retainer attached on the oral side. Three levels of tooth resilience and variable retainer bending stiffness (influenced by retainer type, retainer diameter, and retainer material) were simulated. Applying axial or oblique (45° tilt) loads on a central incisor, the force transmission increased from 2% to 65% with increasing tooth resilience and retainer stiffness. Additionally, a smaller retainer diameter reduced the uniformity of the stress distribution in the bonding interfaces, causing concentrated stress peaks within a small field of the bonding area. An increase in retainer stiffness and in tooth resilience as well as a more oblique load direction all lead to higher overall stress in the adhesive bonding area associated with a higher risk of retainer bonding failure. Therefore, it might be recommended to avoid the use of retainers that are excessively stiff, especially in cases with high tooth resilience. Full article

(This article belongs to the Special Issue Application of Bioengineering to Clinical Orthodontics)

► Show Figures

Figure 1

Figure 1
(a) Schematic illustration of the three-point bending test setup and (b) close-up of the relevant parts with a triple-stranded stainless steel retainer during testing. Full article ">Figure 2
(a) Cross section of tooth 31 with all its components including the two load cases and (b) close-up of the model with the least and most stiff retainer. Full article ">Figure 3
Force transmission (Ftrans) to the adhesive bond of the loaded tooth divided by the applied force (Fres) over bending stiffness (k), normalised to the least stiff retainer (k0, no. 4). Data points are fitted with a function of the type seen in Formula (3). The area of conventional hand-bent retainers is highlighted and enlarged with a green rectangle. Full article ">Figure 4
Maximum degree of utilised adhesive bond capacity plotted for each tooth for five different retainers. The retainers demonstrate the effects of a variation in the RD configuration. The least stiff retainer (k/k0 = 1) corresponds to the solid green curve. An increase in the RD is displayed with the dashed blue curve while the switch to a solid cross section is displayed with the solid blue curve. A further increase in the RD is displayed with the dashed red curve while the configuration that is different to the dashed blue curve is displayed with the solid red curve. The solid blue and red curves also demonstrate the effect of an increase in the RD. All the displayed models correspond to high TR and LC1. Full article ">Figure 5
Degree of utilised capacity for three different retainers. The values were calculated according to the fracture hypothesis as described in Formula (4) (σ/σu + τ/τu), where σ is the normal stress, σu the tensile bond strength, τ the resulting shear stress, and τu the shear bond strength. (a) k/k0 = 1 with RD = 0.38 mm in the multistranded configuration, (b) k/k0 = 12.2 with RD = 0.4 mm and a solid cross section, and (c) k/k0 = 100.1 with RD = 1.20 mm in the multistranded configuration. (a) corresponds to the solid green curve in <a href="#bioengineering-11-00394-f004" class="html-fig">Figure 4</a>, (b) to the solid blue curve, and (c) to the dashed red curve. All displayed models correspond to high TR and LC1. Full article ">Figure 6
Maximum degree of utilised adhesive bond capacity plotted for each tooth for the same retainer but a different load case (dashed line) and a higher TR (blue). Full article ">

19 pages, 2977 KiB

Open AccessArticle

Debulking of the Femoral Stem in a Primary Total Hip Joint Replacement: A Novel Method to Reduce Stress Shielding

by Gulshan Sunavala-Dossabhoy, Brent M. Saba and Kevin J. McCarthy

Bioengineering 2024, 11(4), 393; https://doi.org/10.3390/bioengineering11040393 - 18 Apr 2024

Viewed by 1362

Abstract

In current-generation designs of total primary hip joint replacement, the prostheses are fabricated from alloys. The modulus of elasticity of the alloy is substantially higher than that of the surrounding bone. This discrepancy plays a role in a phenomenon known as stress shielding, [...] Read more.

In current-generation designs of total primary hip joint replacement, the prostheses are fabricated from alloys. The modulus of elasticity of the alloy is substantially higher than that of the surrounding bone. This discrepancy plays a role in a phenomenon known as stress shielding, in which the bone bears a reduced proportion of the applied load. Stress shielding has been implicated in aseptic loosening of the implant which, in turn, results in reduction in the in vivo life of the implant. Rigid implants shield surrounding bone from mechanical loading, and the reduction in skeletal stress necessary to maintain bone mass and density results in accelerated bone loss, the forerunner to implant loosening. Femoral stems of various geometries and surface modifications, materials and material distributions, and porous structures have been investigated to achieve mechanical properties of stems closer to those of bone to mitigate stress shielding. For improved load transfer from implant to femur, the proposed study investigated a strategic debulking effort to impart controlled flexibility while retaining sufficient strength and endurance properties. Using an iterative design process, debulked configurations based on an internal skeletal truss framework were evaluated using finite element analysis. The implant models analyzed were solid; hollow, with a proximal hollowed stem; FB-2A, with thin, curved trusses extending from the central spine; and FB-3B and FB-3C, with thick, flat trusses extending from the central spine in a balanced-truss and a hemi-truss configuration, respectively. As outlined in the International Organization for Standardization (ISO) 7206 standards, implants were offset in natural femur for evaluation of load distribution or potted in testing cylinders for fatigue testing. The commonality across all debulked designs was the minimization of proximal stress shielding compared to conventional solid implants. Stem topography can influence performance, and the truss implants with or without the calcar collar were evaluated. Load sharing was equally effective irrespective of the collar; however, the collar was critical to reducing the stresses in the implant. Whether bonded directly to bone or cemented in the femur, the truss stem was effective at limiting stress shielding. However, a localized increase in maximum principal stress at the proximal lateral junction could adversely affect cement integrity. The controlled accommodation of deformation of the implant wall contributes to the load sharing capability of the truss implant, and for a superior biomechanical performance, the collared stem should be implanted in interference fit. Considering the results of all implant designs, the truss implant model FB-3C was the best model. Full article

(This article belongs to the Special Issue Novel and Advanced Technologies for Orthopaedic Implant)

► Show Figures

Graphical abstract

20 pages, 6271 KiB

Open AccessArticle

Evaluation of Deep Learning Model Architectures for Point-of-Care Ultrasound Diagnostics

by Sofia I. Hernandez Torres, Austin Ruiz, Lawrence Holland, Ryan Ortiz and Eric J. Snider

Bioengineering 2024, 11(4), 392; https://doi.org/10.3390/bioengineering11040392 - 18 Apr 2024

Viewed by 1156

Abstract

Point-of-care ultrasound imaging is a critical tool for patient triage during trauma for diagnosing injuries and prioritizing limited medical evacuation resources. Specifically, an eFAST exam evaluates if there are free fluids in the chest or abdomen but this is only possible if ultrasound [...] Read more.

Point-of-care ultrasound imaging is a critical tool for patient triage during trauma for diagnosing injuries and prioritizing limited medical evacuation resources. Specifically, an eFAST exam evaluates if there are free fluids in the chest or abdomen but this is only possible if ultrasound scans can be accurately interpreted, a challenge in the pre-hospital setting. In this effort, we evaluated the use of artificial intelligent eFAST image interpretation models. Widely used deep learning model architectures were evaluated as well as Bayesian models optimized for six different diagnostic models: pneumothorax (i) B- or (ii) M-mode, hemothorax (iii) B- or (iv) M-mode, (v) pelvic or bladder abdominal hemorrhage and (vi) right upper quadrant abdominal hemorrhage. Models were trained using images captured in 27 swine. Using a leave-one-subject-out training approach, the MobileNetV2 and DarkNet53 models surpassed 85% accuracy for each M-mode scan site. The different B-mode models performed worse with accuracies between 68% and 74% except for the pelvic hemorrhage model, which only reached 62% accuracy for all model architectures. These results highlight which eFAST scan sites can be easily automated with image interpretation models, while other scan sites, such as the bladder hemorrhage model, will require more robust model development or data augmentation to improve performance. With these additional improvements, the skill threshold for ultrasound-based triage can be reduced, thus expanding its utility in the pre-hospital setting. Full article

(This article belongs to the Special Issue Machine-Learning-Driven Medical Image Analysis)

► Show Figures

Figure 1

16 pages, 3827 KiB

Open AccessArticle

Thermal Characterization and Preclinical Feasibility Verification of an Accessible, Carbon Dioxide-Based Cryotherapy System

by Yixin Hu, Naomi Gordon, Katherine Ogg, Dara L. Kraitchman, Nicholas J. Durr and Bailey Surtees

Bioengineering 2024, 11(4), 391; https://doi.org/10.3390/bioengineering11040391 - 18 Apr 2024

Viewed by 1208

Abstract

To investigate the potential of an affordable cryotherapy device for the accessible treatment of breast cancer, the performance of a novel carbon dioxide-based device was evaluated through both benchtop testing and an in vivo canine model. This novel device was quantitatively compared to [...] Read more.

To investigate the potential of an affordable cryotherapy device for the accessible treatment of breast cancer, the performance of a novel carbon dioxide-based device was evaluated through both benchtop testing and an in vivo canine model. This novel device was quantitatively compared to a commercial device that utilizes argon gas as the cryogen. The thermal behavior of each device was characterized through calorimetry and by measuring the temperature profiles of iceballs generated in tissue phantoms. A 45 min treatment in a tissue phantom from the carbon dioxide device produced a 1.67 ± 0.06 cm diameter lethal isotherm that was equivalent to a 7 min treatment from the commercial argon-based device, which produced a 1.53 ± 0.15 cm diameter lethal isotherm. An in vivo treatment was performed with the carbon dioxide-based device in one spontaneously occurring canine mammary mass with two standard 10 min freezes. Following cryotherapy, this mass was surgically resected and analyzed for necrosis margins via histopathology. The histopathology margin of necrosis from the in vivo treatment with the carbon dioxide device at 14 days post-cryoablation was 1.57 cm. While carbon dioxide gas has historically been considered an impractical cryogen due to its low working pressure and high boiling point, this study shows that carbon dioxide-based cryotherapy may be equivalent to conventional argon-based cryotherapy in size of the ablation zone in a standard treatment time. The feasibility of the carbon dioxide device demonstrated in this study is an important step towards bringing accessible breast cancer treatment to women in low-resource settings. Full article

(This article belongs to the Special Issue Novel, Low Cost Technologies for Cancer Diagnostics and Therapeutics)

► Show Figures

Figure 1

15 pages, 2250 KiB

Open AccessArticle

DTONet a Lightweight Model for Melanoma Segmentation

by Shengnan Hao, Hongzan Wang, Rui Chen, Qinping Liao, Zhanlin Ji, Tao Lyu and Li Zhao

Bioengineering 2024, 11(4), 390; https://doi.org/10.3390/bioengineering11040390 - 18 Apr 2024

Viewed by 1329

Abstract

With the further development of neural networks, automatic segmentation techniques for melanoma are becoming increasingly mature, especially under the conditions of abundant hardware resources. This allows for the accuracy of segmentation to be improved by increasing the complexity and computational capacity of the [...] Read more.

With the further development of neural networks, automatic segmentation techniques for melanoma are becoming increasingly mature, especially under the conditions of abundant hardware resources. This allows for the accuracy of segmentation to be improved by increasing the complexity and computational capacity of the model. However, a new problem arises when it comes to actual applications, as there may not be the high-end hardware available, especially in hospitals and among the general public, who may have limited computing resources. In response to this situation, this paper proposes a lightweight deep learning network that can achieve high segmentation accuracy with minimal resource consumption. We introduce a network called DTONet (double-tailed octave network), which was specifically designed for this purpose. Its computational parameter count is only 30,859, which is 1/256th of the mainstream UNet model. Despite its reduced complexity, DTONet demonstrates superior performance in terms of accuracy, with an IOU improvement over other similar models. To validate the generalization capability of this model, we conducted tests on the PH2 dataset, and the results still outperformed existing models. Therefore, the proposed DTONet network exhibits excellent generalization ability and is sufficiently outstanding. Full article

(This article belongs to the Special Issue Application of Deep Learning in Medical Diagnosis)

► Show Figures

Figure 1

Figure 1
The DTO-Net network. (This network consists of an encoding phase and two decoding phases. The encoding phase comprises G_O (ghost octave) and ORFB (octave receptive field block) modules, while the decoding phase consists of the G_ECA (ghost efficient channel attention) module). Full article ">Figure 2
The G_O model (this is the specific composition diagram of the corresponding decoder). Full article ">Figure 3
The G_ECA model (this is the specific composition of encoder component module G_ECA, which is improved on the basis of G_O module to make it more suitable for encoders). Full article ">Figure 4
The ORFB model (this is the specific composition of the encoder component module ORFB, which is improved on the basis of the G_O module, which can improve the accuracy very well, but it will produce too many parameters, so it is only used here). Full article ">Figure 5
Overlayed image of segmentation results and ground truth in ISIC2018. Full article ">Figure 6
Overlay image of segmentation results and ground truth in PH2. Full article ">Figure 7
Comparison of segmentation results of each model. The left part (a) is the presentation of results in the ISIC2018 dataset, and the right part (b) is the presentation of segmentation results in the PH2 dataset. Full article ">

14 pages, 2488 KiB

Open AccessArticle

Altered Functional Connectivity of Temporoparietal Lobe in Obstructive Sleep Apnea: A Resting-State fNIRS Study

by Fang Xiao, Minghui Liu, Yalin Wang, Ligang Zhou, Jingchun Luo, Chen Chen and Wei Chen

Bioengineering 2024, 11(4), 389; https://doi.org/10.3390/bioengineering11040389 - 18 Apr 2024

Viewed by 1484

Abstract

Obstructive Sleep Apnea (OSA), a sleep disorder with high prevalence, is normally accompanied by affective, autonomic, and cognitive abnormalities, and is deemed to be linked to functional brain alterations. To investigate alterations in brain functional connectivity properties in patients with OSA, a comparative [...] Read more.

Obstructive Sleep Apnea (OSA), a sleep disorder with high prevalence, is normally accompanied by affective, autonomic, and cognitive abnormalities, and is deemed to be linked to functional brain alterations. To investigate alterations in brain functional connectivity properties in patients with OSA, a comparative analysis of global and local topological properties of brain networks was conducted between patients with OSA and healthy controls (HCs), utilizing functional near-infrared spectroscopy (fNIRS) imaging. A total of 148 patients with OSA and 150 healthy individuals were involved. Firstly, quantitative alterations in blood oxygen concentration, changes in functional connectivity, and variations in graph theory-based network topological characteristics were assessed. Then, with Mann–Whitney statistics, this study compared whether there are significant differences in the above characteristics between patients with OSA and HCs. Lastly, the study further examined the correlation between the altered characteristics and the apnea hypopnea index (AHI) using linear regression. Results revealed a higher mean and standard deviation of hemoglobin concentration in the superior temporal gyrus among patients with OSA compared to HCs. Resting-state functional connectivity (RSFC) exhibited a slight increase between the superior temporal gyrus and other specific areas in patients with OSA. Notably, neither patients with OSA nor HCs demonstrated significant small-world network properties. Patients with OSA displayed an elevated clustering coefficient (p < 0.05) and local efficiency (p < 0.05). Additionally, patients with OSA exhibited a tendency towards increased nodal betweenness centrality (p < 0.05) and degree centrality (p < 0.05) in the right supramarginal gyrus, as well as a trend towards higher betweenness centrality (p < 0.05) in the right precentral gyrus. The results of multiple linear regressions indicate that the influence of the AHI on RSFC between the right precentral gyrus and right superior temporal gyrus (p < 0.05), as well as between the right precentral gyrus and right supramarginal gyrus (p < 0.05), are statistically significant. These findings suggest that OSA may compromise functional brain connectivity and network topological properties in affected individuals, serving as a potential neurological mechanism underlying the observed abnormalities in brain function associated with OSA. Full article

(This article belongs to the Section Biosignal Processing)

► Show Figures

Figure 1

Figure 1
Two-dimensional view of optode placement. Red denotes the source and blue signifies the detector. Seven sources and seven detectors make up a total of 16 channels. Full article ">Figure 2
The altered brain regions in patients with OSA exhibit distinctive characteristics in mean and variance of HbO2 concentration, with the most prominent changes observed in channel 14, corresponding to the right superior temporal gyrus. Specifically, the mean HbO2 concentration demonstrates a significant difference, with values of 3.81 ± 1.54 compared to 3.65 ± 1.51 in individuals without OSA. Additionally, there is a notable distinction in the standard deviation (std) of HbO2 concentration, showing values of 3.73 ± 1.03 for patients with OSA versus 3.56 ± 1.53 for HCs. * p < 0.05. Full article ">Figure 3
Group-averaged correlation matrix plots of temporoparietal regions for individuals with OSA (a) and HCs (b). Each pixel in the correlation matrix plot represents the z-value of the Pearson correlation coefficient after Fisher z-transformation for the corresponding channel pair. Full article ">Figure 4
Abnormal RSFC within the temporoparietal brain network in individuals with OSA compared to HCs. Full article ">Figure 5
The small-world characteristics of the temporoparietal network for both patients with OSA and HCs across the defined wide threshold range. In the figure, it is evident that the average shortest path length ratio (λ) is approximately equal to one, and the small-world index (σ) is greater than one. However, the clustering coefficient ratio (γ) is not significantly greater than one, indicating that neither patients with OSA nor HCs exhibit a typical small-world topology. Full article ">Figure 6
The between-group differences in global network metrics for the temporoparietal network in individuals with OSA and HCs. There is a noticeable trend suggesting a distinction among patients with OSA, with an elevated Cp (p < 0.05) and an increased Eloc (p < 0.05). * p < 0.05. Full article ">

22 pages, 4938 KiB

Open AccessArticle

Functional and Molecular Analysis of Human Osteoarthritic Chondrocytes Treated with Bone Marrow-Derived MSC-EVs

by Annachiara Scalzone, Clara Sanjurjo-Rodríguez, Rolando Berlinguer-Palmini, Anne M. Dickinson, Elena Jones, Xiao-Nong Wang and Rachel E. Crossland

Bioengineering 2024, 11(4), 388; https://doi.org/10.3390/bioengineering11040388 - 17 Apr 2024

Cited by 1 | Viewed by 2284

Abstract

Osteoarthritis (OA) is a degenerative joint disease, causing impaired mobility. There are currently no effective therapies other than palliative treatment. Mesenchymal stromal cells (MSCs) and their secreted extracellular vesicles (MSC-EVs) have shown promise in attenuating OA progression, promoting chondral regeneration, and modulating joint [...] Read more.

Osteoarthritis (OA) is a degenerative joint disease, causing impaired mobility. There are currently no effective therapies other than palliative treatment. Mesenchymal stromal cells (MSCs) and their secreted extracellular vesicles (MSC-EVs) have shown promise in attenuating OA progression, promoting chondral regeneration, and modulating joint inflammation. However, the precise molecular mechanism of action driving their beneficial effects has not been fully elucidated. In this study, we analyzed MSC-EV-treated human OA chondrocytes (OACs) to assess viability, proliferation, migration, cytokine and catabolic protein expression, and microRNA and mRNA profiles. We observed that MSC-EV-treated OACs displayed increased metabolic activity, proliferation, and migration compared to the controls. They produced decreased proinflammatory (Il-8 and IFN-γ) and increased anti-inflammatory (IL-13) cytokines, and lower levels of MMP13 protein coupled with reduced expression of MMP13 mRNA, as well as negative microRNA regulators of chondrogenesis (miR-145-5p and miR-21-5p). In 3D models, MSC-EV-treated OACs exhibited enhanced chondrogenesis-promoting features (elevated sGAG, ACAN, and aggrecan). MSC-EV treatment also reversed the pathological impact of IL-1β on chondrogenic gene expression and extracellular matrix component (ECM) production. Finally, MSC-EV-treated OACs demonstrated the enhanced expression of genes associated with cartilage function, collagen biosynthesis, and ECM organization and exhibited a signature of 24 differentially expressed microRNAs, associated with chondrogenesis-associated pathways and ECM interactions. In conclusion, our data provide new insights on the potential mechanism of action of MSC-EVs as a treatment option for early-stage OA, including transcriptomic analysis of MSC-EV-treated OA, which may pave the way for more targeted novel therapeutics. Full article

(This article belongs to the Section Regenerative Engineering)

► Show Figures

Figure 1

Figure 1
Characteristics of MSCs and MSC-EVs. (A) MSC morphology and tri-lineage potential (adipogenesis, osteogenesis, chondrogenesis) assessment. (B) FC histogram showing MSC surface phenotype profiles for lineage specific markers (CD14, CD19, CD34, CD45), HLA-DR, CD105, CD73, and CD90. (C) TEM image showing EV morphology. (D FC histogram showing positive expression of EV markers CD63, CD81, and CD9. (E) NTA analysis showing EV size distribution. (F) Western blot analysis showing positive expression of EV markers Flotillin-1 and Alix in three independent MSC-EV examples (S1 to S3). Full article ">Figure 2
MSC-EV-treated OACs’ EV uptake, viability, mobility, and migration. (A) Confocal microscopy images to show MSC-EV-treated OACs. OACs internalize MSC-EVs after 1 h of co-incubation. Cut view confirms internalization of MSC-EVs. (B) Analysis of metabolic activity of OACs and EV-treated OACs (OAC + EV) via CellTiter-Glo 2.0 assay after 5 days of culture. Significance was calculated using the Wilcoxon matched pairs signed-rank test. (C) Two-dimensional scratch assay: example images for OAC and OAC + EV samples at 0 and 40 h post-scratch. Significance was calculated using a mixed-effects model, with Sidak’s multiple comparisons test. (D) Cell migration assay: example images are shown for OAC and OAC + EV samples, assessed using a transwell filter system. Total cell count is shown for OAC and OAC ± EV. Significance was calculated using the paired t-test. (B–D) Lines represent mean ± SEM. The significance threshold for p-values was p < 0.05. Full article ">Figure 3
MSC-EV-treated OACs’ cytokine, catabolic protein, gene, and microRNA expression. (A) Secreted cytokine production (IFN-γ, IL-8, IL-13) by MSC-EV-treated OACs by multiplex immunoassay in n = 12 OAC and OAC + EV. (B) Secreted MMP13 protein production by ELISA in n = 15 OAC and OAC + EV, and gene expression assessed by qRT-PCR in 2D monolayer cultures in n = 11 matched OAC controls and OAC + EV. (C) MicroRNA expression assessed by qRT-PCR in 2D monolayer cultures in n = 11 matched OAC controls and OAC + EV. (A–C) Expression differences between groups were calculated using the paired t-test, and lines represent mean ± SEM. The significance threshold for p-values was p < 0.05. Full article ">Figure 4
Chondrogenic potential of MSC-EV-treated OACs in a 3D spheroid model. (A) Quantification of sGAG production in OAC and OAC + EV samples. (B) Representative histological analysis of toluidine blue, which stains sGAGs, on tissue slices from three independent experiments (scale bar = 200 µm). Zoomed image is shown in the center and in the peripheral zone of the spheroid slice (scale bar = 125 µm). (C) ACAN relative expression via RT-qPCR, for OAC and OAC + EV, at day 21 with respect to day 2. (D) Representative IHC of aggrecan on tissue slices for OAC and OAC + EV samples (scale bar = 200 µm). Zoomed image is shown in the center and in the peripheral zone of the spheroid slice (scale bar = 125 µm). (A,C) Significance was calculated using the paired t-test and lines represent mean ± SEM. The significance threshold for p-values was p < 0.05. Full article ">Figure 5
Effect of MSC-EVs and IL-1β on OAC chondrogenic gene expression and ECM component. (A) Relative gene expression of ACAN, SOX9, COL2A1, COL10A1, MMP13, and ADAMTS at day 10 with respect to day 2. (B) Quantification of sGAG production in digested spheroids formed from OAC, OAC + IL, and OAC + IL + EV samples. (C) Representative histological analysis of Toluidine blue staining GAGs from three independent experiments. Scale bars represent 200 µm and 125 µm, respectively, for original and zoomed-in areas, as indicated. (D) Representative IHC of aggrecan staining on tissue sections of OAC, OAC + IL, and OAC + IL + EV samples from three independent experiments. Scale bars represent 200 µm and 125 µm, respectively, for original and zoomed areas, as indicated. (A,B) Significance was calculated using one-way ANOVA and lines represent mean ± SEM. The significance threshold for p-values was p < 0.05. Full article ">Figure 6
Gene expression profile of MSC-EV-treated OACs compared to OACs. (A) Heatmap shows unsupervised hierarchical clustering of 54 genes that were significantly differentially expressed between groups: each column represents an individual sample. OAC + EV are depicted in blue, while control OACs are depicted in pink. Patient gender (green, purple) and age (green scale) are also indicated. Relative expression changes are indicated by the color scale (red: high; blue: low). (B) Principal component analysis of MSC-EV-treated OACs vs. control OACs. (C) Reactome pathway analysis for genes significantly differentially expressed between OAC + EV and OAC controls. Both analyses indicate the number of genes implicated in each pathway. Full article ">Figure 7
Differential microRNA expression profiles of OAC + EV and OAC. (A) NanoString microRNA expression profiling: heatmap shows unsupervised hierarchical clustering of significantly differentially expressed microRNAs (p < 0.05, n = 24), based on normalized expression counts, in OAC + EV and controls. Each column represents an individual sample. Relative expression changes are indicated by the color scale (red: high; blue: low). OAC + EV are depicted by black shading, while OAC are depicted by grey shading. (B) KEGG pathways analysis for the 24 differentially expressed microRNAs in OAC + EV vs. control OACs. Pathway analysis was performed using Diana Tools miRPath, based on TarBase v8.0 and the number of target genes and targeting microRNAs are plotted for each pathway (FDR-corrected p-values < 0.05). (C) Validation of selected microRNA expression in an independent cohort of OAC + EV and OAC controls (n = 11) by qRT-PCR. Lines represent mean ± SEM. Significance was calculated using the paired t-test, and the threshold for p-values was p < 0.05. Heatmap to show unsupervised hierarchical clustering of the same microRNAs according to NanoString data (n = 7 matched samples and controls). Full article ">

13 pages, 4023 KiB

Open AccessArticle

Could Dental Material Reuse Play a Significant Role in Preservation of Raw Materials, Water, Energy, and Costs? Microbiological Analysis of New versus Reused Healing Abutments: An In Vitro Study

by Roberto Burioni, Lucia Silvestrini, Bianca D’Orto, Giulia Tetè, Matteo Nagni, Elisabetta Polizzi and Enrico Felice Gherlone

Bioengineering 2024, 11(4), 387; https://doi.org/10.3390/bioengineering11040387 - 16 Apr 2024

Cited by 1 | Viewed by 1070

Abstract

Aim: The objective of this in vitro study was to compare reused and sterilized versus new healing abutments to assess whether a decontamination and sterilization process performed on resued healing abutments was sufficient to remove residual proteins. The two groups were comparable with [...] Read more.

Aim: The objective of this in vitro study was to compare reused and sterilized versus new healing abutments to assess whether a decontamination and sterilization process performed on resued healing abutments was sufficient to remove residual proteins. The two groups were comparable with respect to patient safety. Materials and methods: During the period from September 2022 to October 2023, healing abutment screws were selected and divided into two groups according to whether they were new or previously used in patients. The samples were subjected to a decontamination and sterilization protocol, and results from sample sterility evaluation and assessment of surface protein levels were recorded. Results: The obtained results revealed a significant difference in the OD562 nm values between new and reused healing abutment samples. The assay demonstrates how treated healing abutments were still contaminated by residual proteins. Conclusion: Within the limitations of the present study, although from an infectious point of view sterilization results in the total eradication of pathogens, surface proteins remain on reused healing abutments. Full article

(This article belongs to the Special Issue Dental Implant Reconstruction and Biomechanical Evaluation)

► Show Figures

Figure 1

17 pages, 5127 KiB

Open AccessArticle

Balance Evaluation Based on Walking Experiments with Exoskeleton Interference

by Liping Wang, Xin Li, Yiying Peng, Jianda Han and Juanjuan Zhang

Bioengineering 2024, 11(4), 386; https://doi.org/10.3390/bioengineering11040386 - 16 Apr 2024

Viewed by 1153

Abstract

The impairment of walking balance function seriously affects human health and will lead to a significantly increased risk of falling. It is important to assess and improve the walking balance of humans. However, existing evaluation methods for human walking balance are relatively subjective, [...] Read more.

The impairment of walking balance function seriously affects human health and will lead to a significantly increased risk of falling. It is important to assess and improve the walking balance of humans. However, existing evaluation methods for human walking balance are relatively subjective, and the selected metrics lack effectiveness and comprehensiveness. We present a method to construct a comprehensive evaluation index of human walking balance. We used it to generate personal and general indexes. We first pre-selected some preliminary metrics of walking balance based on theoretical analysis. Seven healthy subjects walked with exoskeleton interference on a treadmill at 1.25 m/s while their ground reaction force information and kinematic data were recorded. One subject with Charcot–Marie–Tooth walked at multiple speeds without the exoskeleton while the same data were collected. Then, we picked a number of effective evaluation metrics based on statistical analysis. We finally constructed the Walking Balance Index (WBI) by combining multiple metrics using principal component analysis. The WBI can distinguish walking balance among different subjects and gait conditions, which verifies the effectiveness of our method in evaluating human walking balance. This method can be used to evaluate and further improve the walking balance of humans in subsequent simulations and experiments. Full article

(This article belongs to the Special Issue Bioengineering of the Motor System)

► Show Figures

Figure 1

Figure 1
Gait experimental platform. The gait experimental platform included a control system, an actuation system, transmissions, a motion capture system, and an ankle exoskeleton. Full article ">Figure 2
Three ankle interference torque profiles. These torques were applied only in the support phase (0–60% of a gait cycle) for safety. (A) Sinusoidal interference torque. It consisted of two continuous sinusoids. (B) Constant interference torque. Its maximum torque remained the same. (C) Random interference torque. The torque profile was generated randomly. Full article ">Figure 3
The changes in the preliminary metrics. (A) The changes in the positions and velocities of the COP. (B) The changes in the accelerations of the COM in three directions and the resultant acceleration. (C) The changes in the positions of the COM. (D) The changes in the relative position between the COP and CMP, MOS, and trunk angular acceleration. The blue curves show the changes for subject2 under the ZT condition. The red curves show the changes for subject2 under the MIX condition. The purple curves show the changes for subject8 with CMT under the NW condition. The shadow areas show the fluctuation range of different gait cycles. Full article ">Figure 4
The values of personal WBI for seven healthy subjects. The blue curves are the WBI of each stride under the ZT condition. The red curves and areas are the mean and fluctuation range of the WBI under the MIX condition. Full article ">Figure 5
The results of WBI for subject2. (A) The WBI under ZT and CON conditions (above). The WBI under ZT and RAN conditions (below). (B) Bars and whiskers are means and standard deviations of the WBI for subject2 under ZT and CON conditions (above) and ZT and RAN conditions (below). Full article ">Figure 6
The values of the WBI for the six healthy subjects using the general index constructed from one subject’s data obtained under ZT and MIX conditions. Bars and whiskers are the means and standard deviations of the WBI for six healthy subjects under different gait conditions. Full article ">Figure 7
WBI of healthy subject1 under ZT and MIX conditions and subject8 with CMT under the NW condition. (A) The results of the WBI constructed from subject8’s NW condition and subject1’s MIX condition. (B) The results of the WBI constructed from subject1’s ZT and MIX conditions. The blue curves show subject1’s WBI for each stride under the ZT condition. The red curves and areas show the mean and fluctuation range of subject1’s WBI under the MIX condition. The purple curves and areas show the mean and fluctuation range of subject8’s WBI under the NW condition. Full article ">

11 pages, 1891 KiB

Open AccessEditor’s ChoiceArticle

Application of the Single Source—Detector Separation Algorithm in Wearable Neuroimaging Devices: A Step toward Miniaturized Biosensor for Hypoxia Detection

by Thien Nguyen, Soongho Park, Jinho Park, Asma Sodager, Tony George and Amir Gandjbakhche

Bioengineering 2024, 11(4), 385; https://doi.org/10.3390/bioengineering11040385 - 16 Apr 2024

Viewed by 1513

Abstract

Most currently available wearable devices to noninvasively detect hypoxia use the spatially resolved spectroscopy (SRS) method to calculate cerebral tissue oxygen saturation (StO₂). This study applies the single source—detector separation (SSDS) algorithm to calculate StO₂. Near-infrared spectroscopy (NIRS) data [...] Read more.

Most currently available wearable devices to noninvasively detect hypoxia use the spatially resolved spectroscopy (SRS) method to calculate cerebral tissue oxygen saturation (StO₂). This study applies the single source—detector separation (SSDS) algorithm to calculate StO₂. Near-infrared spectroscopy (NIRS) data were collected from 26 healthy adult volunteers during a breath-holding task using a wearable NIRS device, which included two source—detector separations (SDSs). These data were used to derive oxyhemoglobin (HbO) change and StO₂. In the group analysis, both HbO change and StO₂ exhibited significant change during a breath-holding task. Specifically, they initially decreased to minimums at around 10 s and then steadily increased to maximums, which were significantly greater than baseline levels, at 25–30 s (p-HbO < 0.001 and p-StO₂ < 0.05). However, at an individual level, the SRS method failed to detect changes in cerebral StO₂ in response to a short breath-holding task. Furthermore, the SSDS algorithm is more robust than the SRS method in quantifying change in cerebral StO₂ in response to a breath-holding task. In conclusion, these findings have demonstrated the potential use of the SSDS algorithm in developing a miniaturized wearable biosensor to monitor cerebral StO₂ and detect cerebral hypoxia. Full article

(This article belongs to the Special Issue Neuroimaging Techniques for Wearable Devices in Bioengineering)

► Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
Illustration of devices that use the SRS method and SSDS algorithm. SRS-based devices need to have at least 2 source—detector pairs, while SSDS-based devices need only one source—detector pair; hence, they can be made smaller and simpler. In addition, there is an assumption that d has to be significantly larger than ∂d in the SRS method. As a result, SRS-based devices need to have detectors far away from the light source, which reduces the signal to noise ratio and increases the device size. Full article ">Figure 2
Cerebral HbO changes due to the breath-holding task; (a) HbO change from 3 cm SDS; (b) HbO change from 4 cm SDS. Error bars represent standard error. * indicates a p value < 0.05 and ** indicates a p value < 0.001. Full article ">Figure 3
Change in StO2-SRS due to the breath-holding task. Error bars represent standard error. * indicates a p value < 0.05. StO2-SRS change was obtained from the processed data worksheet. Full article ">Figure 4
Change in StO2-SSDS due to the breath-holding task. (a) StO2-SSDS calculated from 3 cm SDS; (b) StO2-SSDS calculated from 4 cm SDS. Error bars represent standard error. * indicates a p value < 0.05. Full article ">Figure 5
StO2-SRS and StO2-SSDS before, during, and after a breath-holding task; (a) short task and (b) long task. Vertical lines mark the start and end of the breath-holding task. Full article ">

11 pages, 1901 KiB

Open AccessArticle

A Clinical Trial Evaluating the Efficacy of Deep Learning-Based Facial Recognition for Patient Identification in Diverse Hospital Settings

by Ayako Sadahide, Hideki Itoh, Ken Moritou, Hirofumi Kameyama, Ryoya Oda, Hitoshi Tabuchi and Yoshiaki Kiuchi

Bioengineering 2024, 11(4), 384; https://doi.org/10.3390/bioengineering11040384 - 15 Apr 2024

Viewed by 1556

Abstract

Background: Facial recognition systems utilizing deep learning techniques can improve the accuracy of facial recognition technology. However, it remains unclear whether these systems should be available for patient identification in a hospital setting. Methods: We evaluated a facial recognition system using deep learning [...] Read more.

Background: Facial recognition systems utilizing deep learning techniques can improve the accuracy of facial recognition technology. However, it remains unclear whether these systems should be available for patient identification in a hospital setting. Methods: We evaluated a facial recognition system using deep learning and the built-in camera of an iPad to identify patients. We tested the system under different conditions to assess its authentication scores (AS) and determine its efficacy. Our evaluation included 100 patients in four postures: sitting, supine, and lateral positions, with and without masks, and under nighttime sleeping conditions. Results: Our results show that the unmasked certification rate of 99.7% was significantly higher than the masked rate of 90.8% (p < 0.0001). In addition, we found that the authentication rate exceeded 99% even during nighttime sleeping. Furthermore, the facial recognition system was safe and acceptable for patient identification within a hospital environment. Even for patients wearing masks, we achieved a 100% success rate for authentication regardless of illumination if they were sitting with their eyes open. Conclusions: This is the first systematical study to evaluate facial recognition among hospitalized patients under different situations. The facial recognition system using deep learning for patient identification shows promising results, proving its safety and acceptability, especially in hospital settings where accurate patient identification is crucial. Full article

(This article belongs to the Special Issue Artificial Intelligence (AI) in Biomedicine)

► Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
(A) Standard photography; (B) supine position with low illumination; (C) left lateral position with low illumination. Full article ">Figure 2
Authentication scores (AS) in 16 patterns depending on the combination of masking, eyes, body position, and illumination. The AS is shown in the upper half, and the 16 patterns are in the lower half. The AS of intentionally wrong patients was significantly lower than that of correct patients for each situation (p < 0.0001). In the upper panel, red boxes = scores for correct patients, and black boxes = scores for wrong patients. In the lower panel, solid boxes = conditions of mask, closed eyes, supine, and low illumination. Full article ">Figure 3
False rejection rate (FRR) and false acceptance rate (FAR) under conditions (A) without and (B) with a mask. The upper panels show the total range, and the lower panels extend the crossing ranges between the FRR and the FAR. The threshold is the maximum score of wrong matches among all 158,400 values. The thresholds without and with masks were 0.642 and 0.620, respectively. FRR = red lines, FAR = blue lines. Full article ">Figure 4
Frequency of successful matches in 16 combinations of mask-wearing, eye closure, body position, and illumination. The rate of successful matches can be indicated as [(1-FRR) × 100 (%)]. The frequency of successful matches is shown in the upper part, and the 16 patterns are in the lower part. In the lower panel, solid boxes = conditions with mask, closed eyes, supine, sufficient or low illumination. The exact numbers are provided in the top row of the matching rate bar chart. Full article ">Figure 5
Authentication scores (A) and successful matching rates (B) in the nighttime condition. All images were taken under the low illumination condition. In the lower panel, solid boxes = conditions with mask, closed eyes, supine, and low illumination; R = right lateral position; L = left lateral position. In the upper panel in (A), red boxes = scores for correct patients, and black boxes = scores for wrong patients. The exact numbers are provided in the top row of the matching rate bar chart. Full article ">

15 pages, 3732 KiB

Open AccessArticle

Primary Stability of Implants Inserted into Polyurethane Blocks: Micro-CT and Analysis In Vitro

by Chadi Dura Haddad, Ludovica Andreatti, Igor Zelezetsky, Davide Porrelli, Gianluca Turco, Lorenzo Bevilacqua and Michele Maglione

Bioengineering 2024, 11(4), 383; https://doi.org/10.3390/bioengineering11040383 - 15 Apr 2024

Viewed by 1249

Abstract

The approach employed for the site preparation of the dental implant is a variable factor that affects the implant’s primary stability and its ability to integrate with the surrounding bone. The main objective of this in vitro study is to evaluate the influence [...] Read more.

The approach employed for the site preparation of the dental implant is a variable factor that affects the implant’s primary stability and its ability to integrate with the surrounding bone. The main objective of this in vitro study is to evaluate the influence of different techniques used to prepare the implant site on the primary stability of the implant in two different densities of artificial bone. Materials and Methods: A total of 150 implant sites were prepared in rigid polyurethane blocks to simulate two distinct bone densities of 15 pounds per cubic foot (PCF) and 30 PCF, with a 1-mm-thick simulated cortex. The implant sites were equally distributed among piezoelectric surgery (PES), traditional drills (TD), and black ruby magnetic mallet inserts (MM). Two methods have been employed to evaluate the implant’s primary stability, Osstell and micro-tomography. Results: In the present study, we observed significant variations in the implant stability quotient (ISQ) values. More precisely, our findings indicate that the ISQ values were generally higher for 30 PCF compared to 15 PCF. In terms of the preparation technique, PES exhibited the greatest ISQ values, followed by MM, and finally TD. These findings corresponded for both bone densities of 30 PCF (PES 75.6 ± 1.73, MM 69.8 ± 1.91, and TD 65.8 ± 1.91) and 15 PCF (PES 72.3 ± 1.63, MM 62.4 ± 1.77, and TD 60.6 ± 1.81). By utilizing Micro-CT scans, we were able to determine the ratio of the implant occupation to the preparation site. Furthermore, we could calculate the maximum distance between the implant and the wall of the preparation site. The findings demonstrated that PES had a higher ratio of implant to preparation site occupation, followed by TD, and then the MM, at a bone density of 30 PCF (PES 96 ± 1.95, TD 94 ± 1.88, and MM 90.3 ± 2.11). Nevertheless, there were no statistically significant differences in the occupation ratio among these three approaches in the bone density of 15 PCF (PES 89.6 ± 1.22, TD 90 ± 1.31, and MM 88.4 ± 1.17). Regarding the maximum gap between the implant and the site preparation, the smallest gaps were seen when TD were used, followed by MM, and finally by PES, either in a bone density 15 PCF (PES 318 ± 21, TD 238 ± 17, and MM 301 ± 20 μm) or in a bone density 30 PCF (PES 299 ± 20, TD 221 ± 16, and MM 281 ± 19 μm). A statistical analysis using ANOVA revealed these differences to be significant, with p-values of < 0.05. Conclusion: The outcomes of this study indicate that employing the PES technique and osteo-densification with MM during implant insertion may enhance the primary stability and increase the possibility of early implant loading. Full article

► Show Figures

Figure 1

15 pages, 6261 KiB

Open AccessArticle

Limb Volume Measurements: A Comparison of Circumferential Techniques and Optoelectronic Systems against Water Displacement

by Giovanni Farina, Manuela Galli, Leonardo Borsari, Andrea Aliverti, Ioannis Th. Paraskevopoulos and Antonella LoMauro

Bioengineering 2024, 11(4), 382; https://doi.org/10.3390/bioengineering11040382 - 15 Apr 2024

Cited by 1 | Viewed by 1219

Abstract

Background. Accurate measurements of limb volumes are important for clinical reasons. We aimed to assess the reliability and validity of two centimetric and two optoelectronic techniques for limb volume measurements against water volumetry, defined as the gold standard. Methods. Five different measurement methods [...] Read more.

Background. Accurate measurements of limb volumes are important for clinical reasons. We aimed to assess the reliability and validity of two centimetric and two optoelectronic techniques for limb volume measurements against water volumetry, defined as the gold standard. Methods. Five different measurement methods were executed on the same day for each participant, namely water displacement, fixed-height (circumferences measured every 5 (10) cm for the upper (lower limb) centimetric technique, segmental centimetric technique (circumferences measured according to proportional height), optoelectronic plethysmography (OEP, based on a motion analysis system), and IGOODI Gate body scanner technology (which creates an accurate 3D avatar). Results. A population of 22 (15 lower limbs, 11 upper limbs, 8 unilateral upper limb lymphoedema, and 6 unilateral lower limb lymphoedema) participants was selected. Compared to water displacement, the fixed-height centimetric method, the segmental centimetric method, the OEP, and the IGOODI technique resulted in mean errors of 1.2, 0.86, −16.0, and 0.71%, respectively. The corresponding slopes (and regression coefficients) of the linear regression lines were 1.0002 (0.98), 1.0047 (0.99), 0.874 (0.94) and 0.9966 (0.99). Conclusion. The centimetric methods and the IGOODI system are accurate in measuring limb volume with an error of <2%. It is important to evaluate new objective and reliable techniques to improve diagnostic and follow-up possibilities. Full article

(This article belongs to the Special Issue Optical Techniques for Biomedical Engineering)

► Show Figures

Figure 1

19 pages, 1491 KiB

Open AccessArticle

Development of a Human B7-H3-Specific Antibody with Activity against Colorectal Cancer Cells through a Synthetic Nanobody Library

by Jingxian Li, Bingjie Zhou, Shiting Wang, Jiayi Ouyang, Xinyi Jiang, Chenglin Wang, Teng Zhou, Ke-wei Zheng, Junqing Wang and Jiaqi Wang

Bioengineering 2024, 11(4), 381; https://doi.org/10.3390/bioengineering11040381 - 15 Apr 2024

Viewed by 1847

Abstract

Nanobodies have emerged as promising tools in biomedicine due to their single-chain structure and inherent stability. They generally have convex paratopes, which potentially prefer different epitope sites in an antigen compared to traditional antibodies. In this study, a synthetic phage display nanobody library [...] Read more.

Nanobodies have emerged as promising tools in biomedicine due to their single-chain structure and inherent stability. They generally have convex paratopes, which potentially prefer different epitope sites in an antigen compared to traditional antibodies. In this study, a synthetic phage display nanobody library was constructed and used to identify nanobodies targeting a tumor-associated antigen, the human B7-H3 protein. Combining next-generation sequencing and single-clone validation, two nanobodies were identified to specifically bind B7-H3 with medium nanomolar affinities. Further characterization revealed that these two clones targeted a different epitope compared to known B7-H3-specific antibodies, which have been explored in clinical trials. Furthermore, one of the clones, dubbed as A6, exhibited potent antibody-dependent cell-mediated cytotoxicity (ADCC) against a colorectal cancer cell line with an EC50 of 0.67 nM, upon conversion to an Fc-enhanced IgG format. These findings underscore a cost-effective strategy that bypasses the lengthy immunization process, offering potential rapid access to nanobodies targeting unexplored antigenic sites. Full article

(This article belongs to the Section Nanobiotechnology and Biofabrication)

► Show Figures

Figure 1

Figure 1
Quality evaluation of synthetic nanobody library by next-generation sequencing. (A) Diagram depicting the synthetic nanobody gene. The framework regions, highlighted in gray, remained constant in sequence. Conversely, sections of the CDR loops were subject to variation, with CDR1, CDR2, and CDR3 depicted in orange, blue, and green, respectively. Partial randomization was accomplished using mixed nucleotides, while highly variable regions were randomized using degenerate codons NNB, denoted by asterisks (*). (B) For demonstration, amino acid distributions in desired positions were shown for sequences with the short CDR3 in the library. The IMGT definition was used to annotate the CDRs using the AbYsis tool (<a href="http://www.abysis.org/abysis/index.html" target="_blank">http://www.abysis.org/abysis/index.html</a>, accessed on 6 June 2023). (C) The observed and expected ratios for 20 amino acid types over fully randomized sites in nanobodies with the short CDR3 loop. (D) Percentage of intact sequences in NGS data for the E. coli library (E1), three batches of independently prepared phage display libraries without anti-Myc tag enrichment (P1, P2, and P3), and the phage library after anti-Myc tag enrichment (P-Myc). (E) Pie charts show percentages of sequences with one, two, three, and more appearances in corresponding libraries as (D). Full article ">Figure 2
Phage display panning process and NGS mining. (A) The panning process is shown schematically with the amount of recombinant proteins used in each round listed. B7-H3-coated wells are denoted by a “+” sign, and skimmed-milk blocked-control wells are denoted by a “−” sign. The solid lines with arrows indicate that the eluted phages from the previous round were amplified prior to the subsequent round. The dashed line with an arrow indicates that unbound phages from the preceding incubation step were transferred directly to the succeeding wells. In round 3, phages eluted from B7-H3 (−) (negative pool) and B7-H3 (+) (positive pool) were amplified and used for NGS analysis. (B) Single-colony phage ELISA identified potential specific clones. Phages produced from randomly picked colonies were tested for the binding to 4IgB7-H3 or skimmed-milk-blocked control wells. The specificity score was calculated by dividing the absorbance in an antigen-coated well by the absorbance in the blocked control well. Clones with absorbance below 0.1 in the control wells and a specificity score higher than five were determined to be specific binders. Gray and red bars indicated non-specific and specific binders, respectively. (C) NGS result of round 3. The ratio of percentages in the positive versus the negative pools is graphed for each unique sequence against its corresponding percentage in the positive pool. In this analysis, only sequences exceeding 0.01% in the positive pool were considered. Each blue open circle or colored solid circle represented a unique sequence. Full article ">Figure 3
Binding validation for selected clones. (A) ELISA showed binding of selected clones in hFc format to 4IgB7-H3-coated wells and blocked control wells. (B) A6-hFc and a13-hFc showed dose-dependent binding to recombinant B7-H3 in indirect ELISA assay. B7-H3-specific MAb 8H9 was used as the positive control. (C) BLI sensorgrams showed A6, a13, and 8H9 bound to a serial dilution (900 nM, 300 nM, 100 nM, and 33 nM) of recombinant B7-H3. Raw data are in blue, cyan, and green, respectively, and the fitting curves are in red. Full article ">Figure 4
Epitope binning analysis. Streptavidin-coated biosensors were coupled with biotinylated (A) 8H9-hFc or (B) A6-hFc, followed by antigen C-2IgB7-H3 and incubated with the sandwiching MAb (MAb2). The labels indicate the MAb2 identity. (C) Sequence alignment of human and murine B7-H3 with residue numbers provided (referring to the N-terminus of human B7-H3). Residues 1-28 are suggested to be the signal sequence and cleaved during protein maturation. (D) Predicted structure of the N-terminal 2Ig of human B7-H3 was extracted from AlphaFold protein structure database, with non-conserved residues between human and murine B7-H3 depicted in spheres of different colors. Residue numbers (in both N- and C-terminal 2Ig) and the corresponding amino acids in human B7-H3 are labeled. The corresponding amino acids in murine B7-H3 are in parentheses. Full article ">Figure 5
Characterization of A6. (A) Western blot analysis of A6 binding to HCT116 and Jurkat cell lysates, using 8H9 as the positive control. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. (B) Flow cytometric analysis of A6 binding to HCT116 WT and B7-H3 knockdown (B7-H3 KD) cell pools. A human IgG1 isotype was used as a negative control. (C) Left, one representative flow cytometric experiment is shown for the dose-dependent binding of A6-hFc to HCT116. Right, data from three independent experiments were averaged and fitted with a three parameter non-linear model. Full article ">Figure 6
ADCC activity of A6. HCT116 was used as target cells and Jurkat-Lucia NFAT-CD16 (InvivoGen) as reporter cells. The dose-dependent experiment was conducted by measuring the induced luciferase activity of the assay supernatant after 24 h. An IgG1 isotype served as the negative control. The data represent means ± standard deviation (SD) from three independent experiments. Full article ">

14 pages, 3647 KiB

Open AccessArticle

Characteristics of Far-Infrared Ray Emitted from Functional Loess Bio-Balls and Its Effect on Improving Blood Flow

by Yeon Jin Choi, Woo Cheol Choi, Gye Rok Jeon, Jae Ho Kim, Min Seok Kim and Jae Hyung Kim

Bioengineering 2024, 11(4), 380; https://doi.org/10.3390/bioengineering11040380 - 15 Apr 2024

Viewed by 1086

Abstract

XRD diffraction and IR absorption were investigated for raw loess powder and heat-treated loess powder. Raw loess retains its useful minerals, but loses their beneficial properties when calcined at 850 °C and 1050 °C. To utilize the useful minerals, loess balls were made [...] Read more.

XRD diffraction and IR absorption were investigated for raw loess powder and heat-treated loess powder. Raw loess retains its useful minerals, but loses their beneficial properties when calcined at 850 °C and 1050 °C. To utilize the useful minerals, loess balls were made using a low-temperature wet-drying method. The radiant energy and transmittance were measured for the loess balls. Far-infrared ray (FIR) emitted from loess bio-balls is selectively absorbed as higher vibrational energy by water molecules. FIR can raise the body’s core temperature, thereby improving blood flow through the body’s thermoregulatory mechanism. In an exploratory study with 40 participants, when the set temperature of the loess ball mat was increased from 25 °C to 50 °C, blood flow increased by 39.01%, from 37.48 mL/min to 52.11 mL/min, in the left middle finger; in addition, it increased by 39.62%, from 37.15 mL/min to 51.87 mL/min, in the right middle finger. The FIR emitted from loess balls can be widely applied, in various forms, to diseases related to blood flow, such as cold hands and feet, diabetic foot, muscle pain, and menstrual pain. Full article

(This article belongs to the Section Biomedical Engineering and Biomaterials)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Bioengineering, Volume 11, Issue 4 (April 2024) – 112 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI