CN119477863A - A medical ultrasound image quality control system based on multimodal fusion - Google Patents
A medical ultrasound image quality control system based on multimodal fusion Download PDFInfo
- Publication number
- CN119477863A CN119477863A CN202411592743.2A CN202411592743A CN119477863A CN 119477863 A CN119477863 A CN 119477863A CN 202411592743 A CN202411592743 A CN 202411592743A CN 119477863 A CN119477863 A CN 119477863A
- Authority
- CN
- China
- Prior art keywords
- image
- image data
- fusion
- medical ultrasonic
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Computer Graphics (AREA)
- Multimedia (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention discloses a medical ultrasonic image quality control system based on multi-mode fusion, which comprises a data preprocessing module, a dynamic self-adaptive fusion module, an image enhancement module and a visualization module, wherein the data preprocessing module is used for acquiring medical ultrasonic image data of different modes from different crowds in real time and preprocessing the medical ultrasonic image data, the dynamic self-adaptive fusion module is used for dynamically adjusting fusion strategies according to the medical ultrasonic image data acquired in real time and introducing a self-adaptive mechanism to dynamically adjust weights of the different mode data, the image enhancement module combines the medical ultrasonic image data with a three-dimensional anatomical structure of a human body by using an augmented reality technology to generate an intuitive and three-dimensional diagnostic image, the visualization module adopts a dynamic image display mode to present the change process of the diagnostic image in real time, and a complete medical ultrasonic image quality control system is constructed through integrated data preprocessing, dynamic self-adaptive fusion, image enhancement, visualization and other modules, and the system can automatically process and analyze the medical ultrasonic image data, thereby improving the quality control efficiency and accuracy.
Description
Technical Field
The invention relates to a medical ultrasonic image processing system, in particular to a medical ultrasonic image quality control system based on multi-mode fusion.
Background
The medical ultrasonic imaging technology is an advanced medical diagnosis means, which uses ultrasonic waves to penetrate human tissues and reflect and image, so that doctors can clearly observe the structure, morphology and functional state of the tissues, thereby diagnosing and monitoring diseases. The technology is widely applied to a plurality of fields such as prenatal examination, internal organ evaluation, tumor detection and the like. The quality control of medical ultrasonic images is realized by comparing an ultrasonic image with a standard image of a normal crowd through an automatic image analysis technology, and accurately detecting and identifying abnormality or defect in the image so as to ensure that the quality of the medical ultrasonic images meets the diagnosis requirement. In recent years, the application of the multi-mode fusion technology in the medical field is increasingly wide, and the multi-mode fusion technology combines different medical image technologies with clinical data, thereby providing more comprehensive and accurate information support for medical image diagnosis and quality control.
However, since the data of medical ultrasound images originate from a wide population and various modalities, a high degree of diversity and complexity is exhibited, making conventional processing methods difficult to effectively cope with. Meanwhile, the information contained in the ultrasonic image data of various different modes has emphasis on how to fuse the information efficiently and accurately, and a great challenge in the current technical field is formed. In addition, medical ultrasound image data is usually presented in a two-dimensional form, so that visual understanding of a three-dimensional anatomical structure of a human body by a doctor is greatly limited, and accuracy and efficiency of diagnosis are further affected.
Therefore, a medical ultrasound image quality control system based on multi-mode fusion is urgently needed to solve the technical problems in the prior art.
Disclosure of Invention
The invention overcomes the defects of the prior art and provides a medical ultrasonic image quality control system based on multi-mode fusion.
In order to achieve the aim, the medical ultrasonic image quality control system based on multi-mode fusion comprises a data preprocessing module, a dynamic self-adaptive fusion module, an image enhancement module and a visualization module;
the data preprocessing module is used for acquiring medical ultrasonic image data of different modes from different crowds in real time and preprocessing the medical ultrasonic image data;
The dynamic self-adaptive fusion module is used for dynamically adjusting a fusion strategy according to medical ultrasonic image data acquired in real time, and introducing a self-adaptive mechanism to dynamically adjust weights of different mode data;
The image enhancement module combines medical ultrasonic image data with a three-dimensional anatomical structure of a human body by using an augmented reality technology to generate an intuitive and three-dimensional diagnostic image;
The visualization module adopts a dynamic image display mode to display the change process of the diagnostic image in real time.
In a preferred embodiment of the present invention, the preprocessing includes denoising, image resampling, normalization and contrast stretching on the medical ultrasound image data, and the algorithm expression formula of the preprocessing is:
Wherein, I preprocessed (x, y) is the preprocessed image pixel value at the coordinate (x, y), I resamled (x, y) is the image pixel value which is processed by denoising and image resampling at the coordinate (x, y), min (I resampled) is the whole minimum pixel value of the I resampled image, and max (I resampled) is the whole maximum pixel value of the I resampled image;
The preprocessing algorithm expression maps the pixel value of the I resampled image linearly into the [0,255] range so that the contrast of the I resampled image is stretched, the denominator max (I resampled)-min(Iresampled) of the preprocessing algorithm expression stretches the pixel value of the I resampled image to the entire [0,255] range, the numerator I resampled(x,y)-min(Iresampled maps the minimum pixel value of the I resampled image to 0, and the maximum pixel value to 255.
In a preferred embodiment of the invention, the fusion strategy uses a multi-scale decomposition algorithm to decompose medical ultrasonic image data of different modes into a plurality of scale images, the decomposed images are fused according to actual application scenes and data characteristics, the fused images are subjected to multi-scale reconstruction to generate new image data, and an algorithm expression formula of the fusion strategy is as follows:
Wj(I)=DWT(I)
Ifused＝IDWT(Fj)
Wherein I is original image data, W j (I) represents wavelet coefficients on a j-th scale, DWT (I) represents discrete wavelet transform functions, F j represents fused wavelet coefficients, and I A and I B respectively represent medical ultrasonic image data from different modes; And The method comprises the steps of respectively representing wavelet coefficients of medical ultrasonic image data of different modes on a j-th scale, wherein alpha and beta are weighting coefficients, alpha and beta= 1;I fused are satisfied to represent fused image data, and IDWT(s) represents an inverse discrete wavelet transform function.
In a preferred embodiment of the present invention, the adaptive mechanism is introduced based on different crowds, different modalities, actual application scenarios and data features, and is used for dynamically adjusting weights of data of different modalities, and an algorithm expression formula of the adaptive mechanism is as follows:
in the formula, W (F j) represents the self-adaptive weight matrix obtained by calculation; f j represents the wavelet coefficient after fusion; and representing the image data obtained by fusion after the self-adaptive mechanism is introduced.
In a preferred embodiment of the present invention, the image enhancement module uses computer graphics and medical image processing techniques to construct a three-dimensional anatomical model of the patient from medical ultrasound image data or other medical image data of the human body, and uses augmented reality techniques to superimpose virtual medical ultrasound image data onto the human body or anatomical model in the real world to generate an enhanced diagnostic image.
In a preferred embodiment of the present invention, the process expression formula for superimposing virtual medical ultrasound image data onto a human body or anatomical model in the real world using the augmented reality technique is:
wherein M represents a three-dimensional anatomical model, f model (DEG) represents a function for constructing the three-dimensional anatomical model, receives original medical image data as input, and outputs the three-dimensional anatomical model M; The method comprises the steps of acquiring medical ultrasonic image data of different modes of different people without pretreatment in real time by a data preprocessing module, acquiring the medical ultrasonic image data of different modes of different people without pretreatment in real time by the data preprocessing module, acquiring diagnostic images by I enhanced, acquiring a composite function by F AR (), receiving the medical ultrasonic image data of different modes of different people without pretreatment acquired in real time by the data preprocessing module and the image data obtained by fusion after introducing an adaptive mechanism, and outputting the diagnostic images I enhanced.
In a preferred embodiment of the present invention, the visualization module includes an interactive interface, which allows a user to interact with the displayed dynamic image by means of touch, gesture or voice command, so as to implement zooming in, zooming out, rotating, cutting, and labeling and analyzing of the specific area.
In a preferred embodiment of the present invention, a method for using a medical ultrasound image quality control system based on multi-modal fusion is applied to the medical ultrasound image quality control system based on multi-modal fusion described above:
Starting a medical ultrasonic image quality control system based on multi-mode fusion, preprocessing medical ultrasonic image data from different crowds without modes in a mode of denoising, image resampling, normalization and contrast stretching by a data preprocessing module in real time, and linearly mapping pixel values of medical ultrasonic images to the medical ultrasonic images
[0,255] Range;
According to medical ultrasonic image data acquired in real time, the system decomposes medical ultrasonic image data from different modes into a plurality of scale images by using a multi-scale decomposition algorithm, fuses the decomposed images according to actual application scenes and data characteristics, and simultaneously introduces a self-adaptive mechanism to dynamically adjust the weights of the data of different modes in the fusion process so as to realize a dynamic adjustment strategy, and then generates new image data by a scale reconstruction technology;
The system builds a three-dimensional anatomical model of the human body by utilizing computer imaging and medical image processing technology and combining an augmented reality technology, and superimposes virtual medical ultrasonic image data on the human body or the anatomical model in the real world to generate an intuitive and three-dimensional diagnostic image;
The visualization module presents the change process of the diagnostic image by adopting a dynamic image display technology, and a user can interact with the diagnostic image through an interactive interface to realize the amplification, reduction, rotation and cutting of the image and the labeling and analysis of a specific area.
In a preferred embodiment of the invention, a computer device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the system or method when executing the computer program.
In a preferred embodiment of the invention, a computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the steps or methods.
The invention solves the defects existing in the background technology, and has the following beneficial effects:
(1) According to the invention, the data preprocessing module is used for carrying out denoising, image resampling, normalization, contrast stretching and other treatments on medical ultrasonic image data acquired in real time, so that the usability and consistency of the data are improved, and the preprocessing algorithm is used for linearly mapping the image pixel value to the range of [0,255], so that the contrast of the image is enhanced, and a clearer image foundation is provided for subsequent analysis.
(2) The invention adopts a dynamic self-adaptive fusion module, adopts a multi-scale decomposition algorithm to decompose the data of different modes into a plurality of scale images, and fuses according to the actual application scene and the data characteristics. And by introducing a self-adaptive mechanism, the weights of different modal data are dynamically adjusted, so that more accurate and flexible multi-modal data fusion is realized.
(3) According to the invention, the three-dimensional anatomical model is constructed by utilizing computer graphics and medical image processing technology through the image enhancement module, virtual medical ultrasonic image data is superimposed on a human body or anatomical model in the real world by combining the augmented reality technology, and an intuitive and three-dimensional diagnostic image is generated, so that the intuitiveness of the image is improved, and the understanding and judgment of doctors on illness conditions are enhanced.
(4) The invention adopts a dynamic image display mode to display the change process of the diagnostic image in real time through a visualization module. Meanwhile, the interactive interface is provided to allow the user to interact with the image in a touch, gesture or voice command mode, so that the image is enlarged, reduced, rotated and cut, and the specific region is marked and analyzed, and the diagnosis efficiency and accuracy of doctors are greatly improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art;
fig. 1 is a system flow diagram of a preferred embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
As shown in FIG. 1, the medical ultrasonic image quality control system based on multi-mode fusion comprises a data preprocessing module, a dynamic self-adaptive fusion module, an image enhancement module and a visualization module;
The data preprocessing module is used for acquiring medical ultrasonic image data of different modes from different crowds in real time and preprocessing the medical ultrasonic image data;
It should be explained that the distinguishing types of different people include:
Age differences, children, adolescents, adults, elderly people, etc. Different groups of people in different ages have differences in physiological structure, metabolic level and the like, so that medical ultrasonic images have different performances.
Sex differences-male and female differ in physiological structure, hormone level, etc., which also affects the performance of medical ultrasound images.
Disease state-healthy people and people suffering from different diseases have obvious differences in medical ultrasonic images. In addition, even the population suffering from the same disease may show different image characteristics due to factors such as severity of disease, duration of disease and the like.
Genetic background-differences in genetic background can also lead to differences in the performance of a population on medical ultrasound images.
In medical ultrasound imaging, different modalities are often referred to as different types of ultrasound imaging techniques or methods that may provide different information about the internal structure of the human body. Common medical ultrasound imaging modalities include:
B-mode ultrasound (B-mode ultrasound), which is the most common ultrasound imaging modality, is used to display two-dimensional images of the internal structure of the human body.
Doppler ultrasound (Doppler ultrasound) is used to detect blood flow velocity and direction and to evaluate the functional state of organs such as heart, blood vessels, etc.
Three-dimensional ultrasound (3D ultrasound) provides a stereoscopic image of the internal structure of the human body, which helps the doctor to more intuitively understand the pathological condition.
Elastography (Elastography) aids in diagnosing certain diseases, such as breast cancer, by assessing tissue stiffness.
Contrast ultrasound (CEUS), which enhances the Contrast of ultrasound images by intravenous injection of Contrast agents, helps to more clearly show the lesion.
Further, the preprocessing comprises the processing of denoising, image resampling, normalizing and contrast stretching on medical ultrasonic image data, and the arithmetic expression formula of the preprocessing is as follows:
Wherein, I preprocessed (x, y) is the preprocessed image pixel value at the coordinate (x, y), I resamled (x, y) is the image pixel value which is processed by denoising and image resampling at the coordinate (x, y), min (I resampled) is the whole minimum pixel value of the I resampled image, and max (I resampled) is the whole maximum pixel value of the I resampled image;
The pixel value of the I resampled image is linearly mapped to the range of [0,255] by the expression formula of the preprocessing algorithm, so that the contrast of the I resampled image is stretched, and the denominator of the expression formula of the preprocessing algorithm
Max (I resampled)-min(Iresampled) stretches the pixel value of the I resampled image to the full [0,255] range, and numerator I resampled(x,y)-min(Iresampled maps the minimum pixel value of the I resampled image to 0 and the maximum pixel value to 255.
The dynamic self-adaptive fusion module is used for dynamically adjusting a fusion strategy according to medical ultrasonic image data acquired in real time, and introducing a self-adaptive mechanism to dynamically adjust weights of different mode data;
further, the fusion strategy uses a multi-scale decomposition algorithm to decompose medical ultrasonic image data of different modes into a plurality of scale images, the decomposed images are fused according to actual application scenes and data characteristics, the fused images are subjected to multi-scale reconstruction to generate new image data, and the algorithm expression formula of the fusion strategy is as follows:
Wj(I)=DWT(I)
Ifused＝IDWT(Fj)
Wherein I is original image data, W j (I) represents wavelet coefficients on a j-th scale, DWT (I) represents discrete wavelet transform functions, F j represents fused wavelet coefficients, and I A and I B respectively represent medical ultrasonic image data from different modes; And The method comprises the steps of respectively representing wavelet coefficients of medical ultrasonic image data of different modes on a j-th scale, wherein alpha and beta are weighting coefficients, alpha and beta= 1;I fused are satisfied to represent fused image data, and IDWT(s) represents an inverse discrete wavelet transform function.
Further, based on different crowds, different modes, actual application scenes and data characteristics, an adaptive mechanism is introduced for dynamically adjusting the weight of the data of the different modes, and an algorithm expression formula of the adaptive mechanism is as follows:
in the formula, W (F j) represents the self-adaptive weight matrix obtained by calculation; f j represents the wavelet coefficient after fusion; and representing the image data obtained by fusion after the self-adaptive mechanism is introduced.
It should be noted that the specific implementation of the adaptive weighting function W (·) depends on the criteria and algorithm employed. For example, if weights are calculated based on local energy, then W (·) would involve calculating a local energy map of wavelet coefficients and assigning weights according to the energy map. If weights are calculated based on sharpness or contrast, then W (·) would involve calculating statistics such as gradients or standard deviations of wavelet coefficients and assigning weights based on these statistics.
Furthermore, the adaptive mechanism can be applied not only to the multi-scale reconstruction stage but also to the multi-scale fusion stage. For example, adaptive weights are introduced within the fusion function to dynamically adjust the contribution of different modality data in the fusion process. Therefore, the whole fusion strategy is more flexible and robust, and can be better suitable for different application scenes and data characteristics.
The image enhancement module combines medical ultrasonic image data with a three-dimensional anatomical structure of a human body by using an augmented reality technology to generate an intuitive and three-dimensional diagnostic image;
The image enhancement module utilizes computer graphics and medical image processing technology to construct a three-dimensional anatomical model in a patient according to medical ultrasonic image data of a human body or other medical image data, and utilizes an augmented reality technology to superimpose virtual medical ultrasonic image data on the human body or anatomical model in the real world to generate an enhanced diagnostic image.
Further, by using the augmented reality technology, a process expression formula for superimposing virtual medical ultrasound image data onto a human body or an anatomical model in the real world is:
wherein M represents a three-dimensional anatomical model, f model (DEG) represents a function for constructing the three-dimensional anatomical model, receives original medical image data as input, and outputs the three-dimensional anatomical model M; Medical ultrasonic image data of different modes of different crowds which are not preprocessed and are acquired in real time by a data preprocessing module are represented;
I enhanced represents a diagnostic image, F AR (DEG) represents a composite function, and medical ultrasonic image data of different modes of different people which are not preprocessed and acquired in real time by a data preprocessing module and image data obtained by fusion after introducing an adaptive mechanism are taken as input to output a diagnostic image I enhanced.
It should be further explained here that, firstly, the processing of the three-dimensional anatomical model and the medical ultrasound image data is performed using two functions M and M, respectivelyTo express, f model (·) is used to construct a function of the three-dimensional anatomical model, accept raw medical image data (including data of multiple modalities, such as CT, MRI, US, etc.) as input, and output the three-dimensional anatomical model M.
Let f peprocess (·) be used to process the function of the post-medical ultrasound image data, accept the raw medical ultrasound image dataAs input, denoising, contrast enhancement and other treatments are carried out, and the processed medical ultrasonic image data I us is output, wherein the formula expression is as follows:
The overlay process of the augmented reality technique is then refined to multiple steps, such as registration, fusion, and rendering. Let f registration (·) denote a registration function, accept the three-dimensional anatomical model M and the processed medical ultrasound image data I us as input, and output a registered data pair (M registered,Ius), whose formula expression is:
(Mregistered,Ius)＝fregistration(M,Ius)。
Then, let f fusion (·) denote the fusion function, accept the registered data pair (M registered,Ius) as input, fuse the medical ultrasound image into the three-dimensional anatomical model, and output the fused data The formula expression is as follows:
Ifused＝ffusion(Mregeistered,Ius)。
the visualization module adopts a dynamic image display mode to display the change process of the diagnostic image in real time.
The visual module comprises an interactive interface which allows a user to interact with the displayed dynamic image in a touch, gesture or voice command mode, so that the image can be enlarged, reduced, rotated and cut, and the specific region can be marked and analyzed.
A using method of a medical ultrasonic image quality control system based on multi-mode fusion is applied to the medical ultrasonic image quality control system based on multi-mode fusion:
Starting a medical ultrasonic image quality control system based on multi-mode fusion, preprocessing medical ultrasonic image data from different crowds without modes in a mode of denoising, image resampling, normalization and contrast stretching by a data preprocessing module in real time, and linearly mapping pixel values of medical ultrasonic images to the medical ultrasonic images
[0,255] Range;
According to medical ultrasonic image data acquired in real time, the system decomposes medical ultrasonic image data from different modes into a plurality of scale images by using a multi-scale decomposition algorithm, fuses the decomposed images according to actual application scenes and data characteristics, and simultaneously introduces a self-adaptive mechanism to dynamically adjust the weights of the data of different modes in the fusion process so as to realize a dynamic adjustment strategy, and then generates new image data by a scale reconstruction technology;
The system builds a three-dimensional anatomical model of the human body by utilizing computer imaging and medical image processing technology and combining an augmented reality technology, and superimposes virtual medical ultrasonic image data on the human body or the anatomical model in the real world to generate an intuitive and three-dimensional diagnostic image;
The visualization module presents the change process of the diagnostic image by adopting a dynamic image display technology, and a user can interact with the diagnostic image through an interactive interface to realize the amplification, reduction, rotation and cutting of the image and the labeling and analysis of a specific area.
Furthermore, the system supports cloud storage and remote access functions, all processed image data and analysis results can be safely uploaded to a cloud server, authorized users can access and view the image data and analysis results through any networking equipment at any time and any place, and sharing of medical resources and development of remote medical services are promoted.
The system also integrates a data encryption and safe transmission mechanism, ensures the safety and privacy protection of the image data in the transmission and storage processes, and accords with the relevant regulations and standards of medical information safety.
The invention constructs a complete medical ultrasonic image quality control system by integrating modules such as data preprocessing, dynamic self-adaptive fusion, image enhancement, visualization and the like. The system can automatically process and analyze medical ultrasonic image data, and improves the quality control efficiency and accuracy. Meanwhile, the invention also provides computer equipment and a computer readable storage medium, so that the system can operate in different hardware and software environments, and the universality and the practicability of the system are further enhanced.
The above-described preferred embodiments according to the present invention are intended to suggest that, from the above description, various changes and modifications can be made by the person skilled in the art without departing from the scope of the technical idea of the present invention. The technical scope of the present invention is not limited to the description, but must be determined according to the scope of claims.
Claims (10)
1. The medical ultrasonic image quality control system based on multi-mode fusion is characterized by comprising a data preprocessing module, a dynamic self-adaptive fusion module, an image enhancement module and a visualization module;
the data preprocessing module is used for acquiring medical ultrasonic image data of different modes from different crowds in real time and preprocessing the medical ultrasonic image data;
The dynamic self-adaptive fusion module is used for dynamically adjusting a fusion strategy according to medical ultrasonic image data acquired in real time, and introducing a self-adaptive mechanism to dynamically adjust weights of different mode data;
The image enhancement module combines medical ultrasonic image data with a three-dimensional anatomical structure of a human body by using an augmented reality technology to generate an intuitive and three-dimensional diagnostic image;
The visualization module adopts a dynamic image display mode to display the change process of the diagnostic image in real time.
2. The medical ultrasound image quality control system based on multi-modal fusion as set forth in claim 1, wherein the preprocessing includes denoising, image resampling, normalization and contrast stretching of the medical ultrasound image data, and the preprocessing has an algorithm expression formula:
Wherein, I preprocessed (x, y) is the preprocessed image pixel value at the coordinate (x, y), I resamled (x, y) is the image pixel value which is processed by denoising and image resampling at the coordinate (x, y), min (I resampled) is the whole minimum pixel value of the I resampled image, and max (I resampled) is the whole maximum pixel value of the I resampled image;
The preprocessing algorithm expression maps the pixel value of the I resampled image linearly into the [0,255] range so that the contrast of the I resampled image is stretched, the denominator max (I resampled)-min(Iresampled) of the preprocessing algorithm expression stretches the pixel value of the I resampled image to the entire [0,255] range, the numerator I resampled(x,y)-min(Iresampled maps the minimum pixel value of the I resampled image to 0, and the maximum pixel value to 255.
3. The medical ultrasonic image quality control system based on multi-mode fusion according to claim 1, wherein the fusion strategy uses a multi-scale decomposition algorithm to decompose medical ultrasonic image data of different modes into a plurality of scale images, the decomposed images are fused according to actual application scenes and data characteristics, the fused images are subjected to multi-scale reconstruction to generate new image data, and an algorithm expression formula of the fusion strategy is as follows:
Wj(I)=DWT(I)
Ifused＝IDWT(Fj)
Wherein I is original image data, W j (I) represents wavelet coefficients on a j-th scale, DWT (I) represents discrete wavelet transform functions, F j represents fused wavelet coefficients, and I A and I B respectively represent medical ultrasonic image data from different modes; And The method comprises the steps of respectively representing wavelet coefficients of medical ultrasonic image data of different modes on a j-th scale, wherein alpha and beta are weighting coefficients, alpha and beta= 1;I fused are satisfied to represent fused image data, and IDWT(s) represents an inverse discrete wavelet transform function.
4. The medical ultrasonic image quality control system based on multi-mode fusion according to claim 1 and 3, wherein the self-adaptive mechanism is introduced based on different crowds, different modes and actual application scenes and data characteristics and is used for dynamically adjusting the weight of different mode data, and the algorithm expression formula of the self-adaptive mechanism is as follows:
in the formula, W (F j) represents the self-adaptive weight matrix obtained by calculation; f j represents the wavelet coefficient after fusion; and representing the image data obtained by fusion after the self-adaptive mechanism is introduced.
5. The medical ultrasound image quality control system based on multi-modal fusion as set forth in claim 1, wherein the image enhancement module utilizes computer graphics and medical image processing techniques to construct a three-dimensional anatomical model in the patient from medical ultrasound image data of the human body or other medical image data, and utilizes augmented reality techniques to superimpose virtual medical ultrasound image data onto the human body or anatomical model in the real world to generate an enhanced diagnostic image.
6. The multi-modal fusion-based medical ultrasound image quality control system of claim 1 and 5, wherein the process expression formula for superimposing virtual medical ultrasound image data onto a human or anatomical model in the real world using the augmented reality technique is:
wherein M represents a three-dimensional anatomical model, f model (DEG) represents a function for constructing the three-dimensional anatomical model, receives original medical image data as input, and outputs the three-dimensional anatomical model M; The method comprises the steps of acquiring medical ultrasonic image data of different modes of different people without pretreatment in real time by a data preprocessing module, acquiring the medical ultrasonic image data of different modes of different people without pretreatment in real time by the data preprocessing module, acquiring diagnostic images by I enhanced, acquiring a composite function by F AR (), receiving the medical ultrasonic image data of different modes of different people without pretreatment acquired in real time by the data preprocessing module and the image data obtained by fusion after introducing an adaptive mechanism, and outputting the diagnostic images I enhanced.
7. The medical ultrasound image quality control system based on multi-modal fusion as set forth in claim 1, wherein the visualization module comprises an interactive interface allowing a user to interact with the displayed dynamic image by means of touch, gesture or voice command, thereby realizing the zooming in, zooming out, rotation, cutting of the image and labeling and analysis of the specific region.
8. The application method of the medical ultrasonic image quality control system based on the multi-modal fusion is applied to the medical ultrasonic image quality control system based on the multi-modal fusion as set forth in any one of claims 1 to 7, and is characterized in that:
starting a medical ultrasonic image quality control system based on multi-mode fusion, preprocessing medical ultrasonic image data from different crowds in modes of denoising, image resampling, normalization and contrast stretching by a data preprocessing module, and linearly mapping pixel values of medical ultrasonic images into a range of [0,255 ];
According to medical ultrasonic image data acquired in real time, the system decomposes medical ultrasonic image data from different modes into a plurality of scale images by using a multi-scale decomposition algorithm, fuses the decomposed images according to actual application scenes and data characteristics, and simultaneously introduces a self-adaptive mechanism to dynamically adjust the weights of the data of different modes in the fusion process so as to realize a dynamic adjustment strategy, and then generates new image data by a scale reconstruction technology;
The system builds a three-dimensional anatomical model of the human body by utilizing computer imaging and medical image processing technology and combining an augmented reality technology, and superimposes virtual medical ultrasonic image data on the human body or the anatomical model in the real world to generate an intuitive and three-dimensional diagnostic image;
The visualization module presents the change process of the diagnostic image by adopting a dynamic image display technology, and a user can interact with the diagnostic image through an interactive interface to realize the amplification, reduction, rotation and cutting of the image and the labeling and analysis of a specific area.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the system or method according to any one of claims 1 to 8 when the computer program is executed.
10. A computer readable storage medium having stored thereon a computer program, which when executed by a processor realizes the steps of any of claims 1 to 8 or the steps of the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411592743.2A CN119477863A (en) | 2024-11-08 | 2024-11-08 | A medical ultrasound image quality control system based on multimodal fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411592743.2A CN119477863A (en) | 2024-11-08 | 2024-11-08 | A medical ultrasound image quality control system based on multimodal fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN119477863A true CN119477863A (en) | 2025-02-18 |
Family
ID=94578651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411592743.2A Pending CN119477863A (en) | 2024-11-08 | 2024-11-08 | A medical ultrasound image quality control system based on multimodal fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119477863A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120257152A (en) * | 2025-03-20 | 2025-07-04 | 启东市妇幼保健院 | A dynamic evaluation method and system for ultrasonic quality control |
| CN120318240A (en) * | 2025-06-18 | 2025-07-15 | 北京凯普顿医药科技开发有限公司 | A method for detecting medical image quality |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103371870A (en) * | 2013-07-16 | 2013-10-30 | 深圳先进技术研究院 | Multimode image based surgical operation navigation system |
| CN104274247A (en) * | 2014-10-20 | 2015-01-14 | 上海电机学院 | Medical surgical navigation method |
| CN108324246A (en) * | 2018-01-19 | 2018-07-27 | 上海联影医疗科技有限公司 | Medical diagnosis auxiliary system and method |
| CN118078442A (en) * | 2024-04-16 | 2024-05-28 | 上海英诗拜科技有限公司 | A surgical AR-assisted method based on ultrasound and CT scanning |
| CN118196074A (en) * | 2024-04-16 | 2024-06-14 | 山东佰泰丰信息科技有限公司 | A medical ultrasound image quality control system based on multimodal fusion |
-
2024
- 2024-11-08 CN CN202411592743.2A patent/CN119477863A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103371870A (en) * | 2013-07-16 | 2013-10-30 | 深圳先进技术研究院 | Multimode image based surgical operation navigation system |
| CN104274247A (en) * | 2014-10-20 | 2015-01-14 | 上海电机学院 | Medical surgical navigation method |
| CN108324246A (en) * | 2018-01-19 | 2018-07-27 | 上海联影医疗科技有限公司 | Medical diagnosis auxiliary system and method |
| CN118078442A (en) * | 2024-04-16 | 2024-05-28 | 上海英诗拜科技有限公司 | A surgical AR-assisted method based on ultrasound and CT scanning |
| CN118196074A (en) * | 2024-04-16 | 2024-06-14 | 山东佰泰丰信息科技有限公司 | A medical ultrasound image quality control system based on multimodal fusion |
Non-Patent Citations (2)
| Title |
|---|
| 暖仔会飞: "Opencv学习之：将图片的值进行范围调整cv2.normalize()", Retrieved from the Internet <URL:https://blog.csdn.net/qq_42902997/article/details/122659376> * |
| 李健等: "基于小波变换的多模态医学图像融合", 中国医学装备, vol. 14, no. 11, 30 November 2017 (2017-11-30), pages 1 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120257152A (en) * | 2025-03-20 | 2025-07-04 | 启东市妇幼保健院 | A dynamic evaluation method and system for ultrasonic quality control |
| CN120318240A (en) * | 2025-06-18 | 2025-07-15 | 北京凯普顿医药科技开发有限公司 | A method for detecting medical image quality |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Han et al. | Liver segmentation with 2.5 D perpendicular UNets | |
| US6983063B1 (en) | Computer-aided diagnosis method for aiding diagnosis of three dimensional digital image data | |
| US11468570B2 (en) | Method and system for acquiring status of strain and stress of a vessel wall | |
| CN110522516A (en) | A multi-level interactive visualization method for surgical navigation | |
| AU2013317201A1 (en) | Device and method for displaying three-dimensional image, and program | |
| US20210319880A1 (en) | Diagnostic Support System and a Diagnostic Support Method | |
| CN101849813A (en) | 3D Cardiac Ultrasound Virtual Endoscopy System | |
| CN114170118B (en) | Semi-supervised multimodal MRI image synthesis method based on coarse-to-fine learning | |
| CN111369675B (en) | Method and device for three-dimensional visual model reconstruction based on visceral pleura projection of pulmonary nodules | |
| KR20210098381A (en) | Device and method for visualizing image of lesion | |
| Abou El-Seoud et al. | An interactive mixed reality ray tracing rendering mobile application of medical data in minimally invasive surgeries | |
| CN114708283A (en) | Image object segmentation method and device, electronic equipment and storage medium | |
| Bucioli et al. | Holographic real time 3D heart visualization from coronary tomography for multi-place medical diagnostics | |
| CN119477863A (en) | A medical ultrasound image quality control system based on multimodal fusion | |
| US20220148162A1 (en) | Method and product for ai endoscope analyzing of vein based on vrds 4d medical images | |
| JP2008509773A (en) | Flexible 3D rotational angiography-computed tomography fusion method | |
| KR101657285B1 (en) | Ultrasonography simulation system | |
| CN119445004B (en) | Liver image three-dimensional reconstruction system and method based on artificial intelligence technology | |
| Khedir et al. | Brainar: Automated brain tumor diagnosis with deep learning and 3d augmented reality visualization | |
| CN106097422A (en) | Liver 3-D view dynamic demonstration system based on big data | |
| CN119150993A (en) | Multi-mode generation model training method and device, text generation method and device | |
| Giraud et al. | A three-dimensional mirror augmented by medical imaging: questioning self-portraying at the limit of iintimacy | |
| US12499552B2 (en) | Image processing apparatus, image processing method, and recording medium | |
| CN117974832A (en) | Multimodal liver medical image augmentation algorithm based on generative adversarial network | |
| US20230039065A1 (en) | System and method for generating three dimensional geometric models of anatomical regions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20250218 |