[go: up one dir, main page]

CN119564343B - Method and system for photoacoustic microscopy imaging of brain electrode implantation sites and path planning - Google Patents

Method and system for photoacoustic microscopy imaging of brain electrode implantation sites and path planning

Info

Publication number
CN119564343B
CN119564343B CN202411761971.8A CN202411761971A CN119564343B CN 119564343 B CN119564343 B CN 119564343B CN 202411761971 A CN202411761971 A CN 202411761971A CN 119564343 B CN119564343 B CN 119564343B
Authority
CN
China
Prior art keywords
image
implantation
photoacoustic
information
dimensional
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.)
Active
Application number
CN202411761971.8A
Other languages
Chinese (zh)
Other versions
CN119564343A (en
Inventor
杨广中
郭遥
柳宇轩
罗雅婷
刘宁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Fuyi Medical Technology Co ltd
Shanghai Jiao Tong University
Original Assignee
Shanghai Fuyi Medical Technology Co ltd
Shanghai Jiao Tong University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Fuyi Medical Technology Co ltd, Shanghai Jiao Tong University filed Critical Shanghai Fuyi Medical Technology Co ltd
Priority to CN202411761971.8A priority Critical patent/CN119564343B/en
Publication of CN119564343A publication Critical patent/CN119564343A/en
Application granted granted Critical
Publication of CN119564343B publication Critical patent/CN119564343B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10Computer-aided planning, simulation or modelling of surgical operations
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/20Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/0033Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room
    • A61B5/004Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part
    • A61B5/0042Features or image-related aspects of imaging apparatus, e.g. for MRI, optical tomography or impedance tomography apparatus; Arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part for the brain
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/0093Detecting, measuring or recording by applying one single type of energy and measuring its conversion into another type of energy
    • A61B5/0095Detecting, measuring or recording by applying one single type of energy and measuring its conversion into another type of energy by applying light and detecting acoustic waves, i.e. photoacoustic measurements
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/25Bioelectric electrodes therefor
    • A61B5/279Bioelectric electrodes therefor specially adapted for particular uses
    • A61B5/291Bioelectric electrodes therefor specially adapted for particular uses for electroencephalography [EEG]
    • A61B5/293Invasive
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/25Bioelectric electrodes therefor
    • A61B5/279Bioelectric electrodes therefor specially adapted for particular uses
    • A61B5/294Bioelectric electrodes therefor specially adapted for particular uses for nerve conduction study [NCS]
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • A61B5/7264Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • A61B5/7264Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A61B5/7267Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/101Computer-aided simulation of surgical operations
    • A61B2034/105Modelling of the patient, e.g. for ligaments or bones
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/107Visualisation of planned trajectories or target regions
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/108Computer aided selection or customisation of medical implants or cutting guides
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/20Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
    • A61B2034/2046Tracking techniques
    • A61B2034/2055Optical tracking systems
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/20Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
    • A61B2034/2046Tracking techniques
    • A61B2034/2063Acoustic tracking systems, e.g. using ultrasound

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Surgery (AREA)
  • Physics & Mathematics (AREA)
  • Veterinary Medicine (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Pathology (AREA)
  • Biophysics (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physiology (AREA)
  • Psychiatry (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Evolutionary Computation (AREA)
  • Radiology & Medical Imaging (AREA)
  • Robotics (AREA)
  • Neurology (AREA)
  • Acoustics & Sound (AREA)
  • Image Processing (AREA)

Abstract

The invention provides a method and a system for planning an implantation site and a path of a brain electrode by using photoacoustic microscopic imaging, wherein the method comprises the steps of extracting an input photoacoustic microscopic scanning imaging result by using a cerebral blood vessel enhancement extraction algorithm to obtain cerebral blood vessel information; the cerebral vascular information comprises two-dimensional image information and a three-dimensional reconstruction result of cerebral vascular, and the implantation site and path acquisition step comprises the step of processing cerebral vascular information by using a cerebral electrode implantation site path planning algorithm to obtain an implantation site on a two-dimensional image and an implantation path in a three-dimensional structure. The invention solves the problems of large imaging noise and poor imaging quality of photoacoustic scanning by adopting a cerebral vessel enhancement extraction algorithm, and realizes accurate extraction and three-dimensional reconstruction of cerebral vessels.

Description

Photoacoustic microscopic imaging brain electrode implantation site and method and system for planning path thereof
Technical Field
The invention relates to the technical field of medical image processing, in particular to a photoacoustic microscopic imaging brain electrode implantation site and a method and a system for planning a path of the photoacoustic microscopic imaging brain electrode implantation site.
Background
Electrode implantation techniques include single needle electrode implants and multi-needle electrode implants (e.g., two electrodes or four electrodes at a time). Compared with a single-needle electrode, the multi-needle electrode can generally obtain a signal with higher quality, and is an electrode implantation mode commonly used in the current animal experiment process. In practice, it is often necessary to rationally plan the implantation site (or implant site) of the electrode prior to performing the electrode implantation procedure. Taking brain electrode implantation as an example, compared with scalp electrodes, nerve electrodes implanted on the cerebral cortex can record nerve electrophysiological signals with higher signal to noise ratio, and can help researchers to understand brain activity changes more effectively. The cerebral cortex has a large number of blood vessels, and if the implantation position is not planned accurately enough, the electrodes can be implanted into the blood vessel area, so that bleeding hidden danger is caused. Therefore, it is important to ensure the accuracy of the planned electrode implantation site.
Photoacoustic microscopic scanning imaging is a preoperative imaging result realized based on photoacoustic microscopic imaging (PhotoAcoustic Microscopy, PAM), the resolution of the photoacoustic microscopic scanning imaging can reach tens to tens of micrometers, and the photoacoustic microscopic scanning imaging can image superficial blood vessels and deep blood vessels, is a novel biomedical imaging means and is commonly used for tissue imaging of blood vessels and the like. Under the excitation of 532nm laser, the cerebral blood vessel has stronger contrast with surrounding brain tissue, and can realize the cerebral blood vessel imaging with high contrast and high signal-to-noise ratio.
However, photoacoustic scan imaging is noisy and has unsatisfactory imaging quality, which may lead to bleeding during surgery and even damage to brain function.
In China patent document with publication number of CN116196097A, an electrode implantation site planning method, an electrode implantation site planning device, a readable storage medium and a terminal are disclosed, wherein the method comprises the steps of determining an original image acquired for an electrode implantation object, dividing the original image to determine a non-implantation area and an implantable area, determining a plurality of first implantation sites in the implantable area based on a first preset electrode distance, determining a preliminary allocation point of each first implantation site based on the first preset electrode distance for each first implantation site, screening each first implantation site and the preliminary allocation point thereof based on a second preset electrode distance, and determining an electrode implantation site planning result. The patent document uses a traditional imaging means, has low spatial resolution, cannot capture all blood vessel information, only considers site planning on a two-dimensional image plane, and does not consider spatial path planning on a three-dimensional plane.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a photoacoustic microscopic imaging brain electrode implantation site and a method and a system for planning a path of the photoacoustic microscopic imaging brain electrode implantation site.
The invention provides a photoacoustic microscopic imaging brain electrode implantation site and a path planning method thereof, comprising the following steps:
Extracting an input photoacoustic microscopic scanning imaging result by using a cerebral blood vessel enhancement extraction algorithm to obtain cerebral blood vessel information, wherein the cerebral blood vessel information comprises two-dimensional image information and a three-dimensional reconstruction result of cerebral blood vessels;
and obtaining implantation sites and paths, namely processing cerebral vessel information by using a cerebral electrode implantation site path planning algorithm to obtain implantation sites on a two-dimensional image and implantation paths in a three-dimensional structure.
Preferably, the image extraction step includes:
S1.1, denoising the acquired photoacoustic microscopic image by using a trained denoising model, and improving the signal-to-noise ratio;
S1.2, enhancing the photoacoustic microscopy image by using a trained image enhancement model, correcting an imaging fuzzy region, a region with resolution not reaching a preset value and an imaging deformation region, and improving the resolution of the photoacoustic microscopy image;
S1.3, segmenting blood vessels in the photoacoustic microscopy image by using a trained blood vessel segmentation model, and removing background and other tissue structure information to obtain two-dimensional image information of cerebral blood vessels;
And S1.4, combining the depth information of the photoacoustic microscopic imaging with the two-dimensional image information of the cerebral blood vessel, and carrying out back projection calculation to obtain a three-dimensional reconstruction result of the cerebral blood vessel.
Preferably, the denoising model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting a noise-removed signal-to-noise ratio photoacoustic microscopic image, the training of the deep neural network is based on supervised learning, and the supervision information is a denoising and noise adding process.
Preferably, the image enhancement model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting the photoacoustic microscopic image with enhanced resolution, the training of the deep neural network is based on supervised learning, and the supervision information is the recovery difference between the original high-resolution image and the downsampled image.
Preferably, the blood vessel segmentation model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting a mask image corresponding to a blood vessel region, the training of the deep neural network is based on supervised learning, and the supervision information is rough segmentation labeling of blood vessels.
Preferably, the implantation point and path obtaining step includes:
S2.1, calculating a two-dimensional cerebrovascular probability density map to obtain implantation sites, selecting starting points for setting the number of implantation sites in a two-dimensional image plane to ensure that the distance between every two implantation sites is more than or equal to the shortest distance between the appointed adjacent sites, and optimizing the target to ensure that the sum of cumulative probability values of all implantation sites is maximum, wherein the optimizing condition is that the distance between every two implantation sites is more than or equal to the shortest distance between the adjacent sites;
And S2.2, calculating a three-dimensional brain vessel probability density map to acquire an implantation path, selecting an implantation direction with the determined implantation site as a starting point in a three-dimensional space, and optimizing the target to maximize the sum of cumulative probability values on all implantation paths under the condition of designating implantation depth.
The invention provides a photoacoustic microscopic imaging brain electrode implantation site and a path planning system thereof, comprising:
The image extraction module is used for extracting an input photoacoustic microscopic scanning imaging result by using a cerebrovascular enhancement extraction algorithm to obtain cerebrovascular information, wherein the cerebrovascular information comprises two-dimensional image information and a three-dimensional reconstruction result of the cerebral blood vessel;
And the implantation site and path acquisition module is used for processing the cerebrovascular information by using a brain electrode implantation site path planning algorithm to obtain an implantation site on a two-dimensional image and an implantation path in a three-dimensional structure.
Preferably, the image extraction module includes:
The module M1.1 uses the trained denoising model to denoise the acquired photoacoustic microscopic image, and improves the signal to noise ratio;
the trained image enhancement model is used for enhancing the photoacoustic microscopy image, the imaging fuzzy area, the area with the resolution not reaching the preset value and the imaging deformation area are corrected, and the resolution of the photoacoustic microscopy image is improved;
the module M1.3 is used for dividing blood vessels in the photoacoustic microscopy image by using a trained blood vessel division model, removing background and other tissue structure information, and obtaining two-dimensional image information of cerebral blood vessels;
And a module M1.4, combining the depth information of the photoacoustic microscopic imaging with the two-dimensional image information of the cerebral blood vessel, and obtaining a three-dimensional reconstruction result of the cerebral blood vessel through back projection calculation.
Preferably, the denoising model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting a noise-removed signal-to-noise ratio photoacoustic microscopic image, the training of the deep neural network is based on supervised learning, and the supervision information is a denoising and noise adding process.
Preferably, the image enhancement model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting the photoacoustic microscopic image with enhanced resolution, the training of the deep neural network is based on supervised learning, and the supervision information is the recovery difference between the original high-resolution image and the downsampled image.
Preferably, the blood vessel segmentation model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting a mask image corresponding to a blood vessel region, the training of the deep neural network is based on supervised learning, and the supervision information is rough segmentation labeling of blood vessels.
Preferably, the implantation point and path acquisition module includes:
Selecting starting points for setting the number of implantation sites in a two-dimensional image plane to ensure that the distance between every two implantation sites is more than or equal to the shortest distance between the appointed adjacent sites, optimizing the target to ensure that the sum of the cumulative probability values of all implantation sites is maximum, and optimizing the target to ensure that the distance between every two sites is more than or equal to the shortest distance between the adjacent sites;
and a module M2.2, calculating a three-dimensional cerebrovascular probability density map to acquire an implantation path, selecting an implantation direction with the determined implantation site as a starting point in a three-dimensional space, and optimizing the target to maximize the sum of cumulative probability values on all implantation paths under the condition of designating implantation depth.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention solves the problems of large imaging noise and poor imaging quality of photoacoustic scanning by adopting a cerebral vessel enhancement extraction algorithm, and realizes accurate extraction and three-dimensional reconstruction of cerebral vessels.
2. The invention is based on the extracted cerebral blood vessel and three-dimensional reconstruction, and realizes the avoidance of the blood vessel in the cerebral electrode implantation process by adopting a cerebral electrode implantation site path planning algorithm, thereby reducing the bleeding and the potential damage to brain functions in the operation process, obtaining the effect of improving the operation safety and success rate and providing great help for doctors.
Other advantages of the present invention will be set forth in the description of specific technical features and solutions, by which those skilled in the art should understand the advantages that the technical features and solutions bring.
Drawings
Other features, objects and advantages of the present invention will become more apparent upon reading of the detailed description of non-limiting embodiments, given with reference to the accompanying drawings in which:
FIG. 1 is a flow chart of the method of the present invention.
Fig. 2 is a flowchart of the cerebrovascular enhancement extraction algorithm of the present invention.
Fig. 3 is a flowchart of a brain electrode implantation site path planning algorithm according to the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the present invention, but are not intended to limit the invention in any way. It should be noted that variations and modifications could be made by those skilled in the art without departing from the inventive concept. These are all within the scope of the present invention.
Referring to fig. 1, a method for photoacoustic microscopy imaging brain electrode implantation site and path planning thereof includes:
Extracting an input photoacoustic microscopic scanning imaging result by using a cerebral blood vessel enhancement extraction algorithm to obtain cerebral blood vessel information, wherein the cerebral blood vessel information comprises two-dimensional image information and a three-dimensional reconstruction result of cerebral blood vessels;
Specific:
S1.1, denoising the acquired photoacoustic microscopic image by using a trained denoising model, and improving the signal-to-noise ratio;
S1.2, enhancing the photoacoustic microscopy image by using a trained image enhancement model, correcting an imaging fuzzy region, a region with resolution not reaching a preset value and an imaging deformation region, and improving the resolution of the photoacoustic microscopy image;
S1.3, segmenting blood vessels in the photoacoustic microscopy image by using a trained blood vessel segmentation model, and removing background and other tissue structure information to obtain two-dimensional image information of cerebral blood vessels;
And S1.4, combining the depth information of the photoacoustic microscopic imaging with the two-dimensional image information of the cerebral blood vessel, and carrying out back projection calculation to obtain a three-dimensional reconstruction result of the cerebral blood vessel.
The denoising model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting a noise ratio photoacoustic microscopic image for removing noise, training of the deep neural network is based on supervised learning, and the supervision information is a denoising and noise adding process.
The image enhancement model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting the photoacoustic microscopic image with enhanced resolution, training of the deep neural network is based on supervised learning, and the supervision information is the recovery difference between the original high-resolution image and the downsampled image.
The blood vessel segmentation model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting a mask image corresponding to a blood vessel region, the training of the deep neural network is based on supervised learning, and the supervision information is rough segmentation labeling of blood vessels.
And obtaining implantation sites and paths, namely processing cerebral vessel information by using a cerebral electrode implantation site path planning algorithm to obtain implantation sites on a two-dimensional image and implantation paths in a three-dimensional structure.
Specific:
S2.1, calculating a two-dimensional cerebrovascular probability density map to obtain implantation sites, selecting starting points for setting the number of implantation sites in a two-dimensional image plane to ensure that the distance between every two implantation sites is more than or equal to the shortest distance between the appointed adjacent sites, and optimizing the target to ensure that the sum of cumulative probability values of all implantation sites is maximum, wherein the optimizing condition is that the distance between every two implantation sites is more than or equal to the shortest distance between the adjacent sites;
And S2.2, calculating a three-dimensional brain vessel probability density map to acquire an implantation path, selecting an implantation direction with the determined implantation site as a starting point in a three-dimensional space, and optimizing the target to maximize the sum of cumulative probability values on all implantation paths under the condition of designating implantation depth.
The invention solves the problems of large imaging noise and poor imaging quality of photoacoustic scanning by adopting a cerebral vessel enhancement extraction algorithm, and realizes accurate extraction and three-dimensional reconstruction of cerebral vessels. Based on the extracted cerebral blood vessels and three-dimensional reconstruction, by adopting a cerebral electrode implantation site path planning algorithm, the blood vessels are avoided in the cerebral electrode implantation process, the bleeding and the potential damage to brain functions in the operation process are reduced, and the effect of improving the operation safety and success rate is achieved.
The foregoing is a basic embodiment of the present invention, and a further description of the technical solution of the present invention is provided below by means of a preferred embodiment.
Example 1
Referring to fig. 1, a method for photoacoustic microscopy imaging brain electrode implantation site and path planning thereof includes:
And S1, executing a cerebral blood vessel enhancement extraction algorithm for an input photoacoustic microscopic scanning imaging result to extract two-dimensional image information and three-dimensional reconstruction results of cerebral blood vessels.
Referring to fig. 2, the process of the cerebral blood vessel enhancement extraction algorithm is that firstly, denoising operation is carried out on an original acquired photoacoustic microscopic image, noise caused by background interference, instrument electromagnetic radiation, imaging tissue motion and the like is removed by using a trained denoising model, and the signal-to-noise ratio of the photoacoustic microscopic image is improved. And then, the trained image enhancement model is utilized to enhance the photoacoustic microscopy image, the imaging fuzzy region, the region with lower resolution and the imaging deformation region are corrected, and the resolution of the photoacoustic microscopy image is improved. Further, the trained blood vessel segmentation model is utilized to segment blood vessels in the photoacoustic microscopy image, and background and other tissue structure information are removed, so that two-dimensional image information of cerebral blood vessels is obtained. Finally, combining the depth information of the photoacoustic microscopic imaging with the two-dimensional image information of the cerebral vessels obtained in the previous step, and performing back projection calculation to obtain three-dimensional reconstruction results of the cerebral vessels.
The denoising model is realized based on a deep neural network, and has the functions of inputting an original photoacoustic microscopic image and outputting a photoacoustic microscopic image with high signal to noise ratio for removing noise. The training of the deep neural network is based on supervised learning, and the supervision information is a denoising and noise adding process.
The image enhancement model is realized based on a deep neural network, and has the functions of inputting an original photoacoustic microscopic image and outputting an enhanced-resolution photoacoustic microscopic image. The training of the deep neural network is based on supervised learning, and the supervision information is the recovery difference between the original high-resolution image and the downsampled image.
The blood vessel segmentation model is realized based on a deep neural network, and has the functions of inputting an original photoacoustic microscopic image and outputting a mask image corresponding to a blood vessel region. The training of the deep neural network is based on supervised learning, and the supervision information is rough segmentation labeling of blood vessels.
And S2, for the extracted cerebrovascular information, executing a brain electrode implantation site path planning algorithm to obtain an implantation site on a two-dimensional image and an implantation path in a three-dimensional structure.
The selection of the implantation site is calculated based on a two-dimensional brain vessel probability density map. The main calculation method is to select starting points for setting the number of implantation sites in a two-dimensional image plane so as to ensure that the distance between every two implantation sites is more than or equal to the shortest distance between the appointed adjacent sites, the optimization target is to ensure that the sum of the cumulative probability values of all implantation sites is maximum, and the optimization condition is to keep that the distance between every two implantation sites is more than or equal to the shortest distance between the adjacent sites.
The implantation path is selected based on a three-dimensional brain vessel probability density map. The main calculation method is to select an implantation direction taking an implantation site as a starting point in a three-dimensional space, and under the condition of designating implantation depth, the optimization target is to maximize the sum of cumulative probability values on all implantation paths.
Referring to fig. 3, the brain electrode implantation site path planning algorithm includes implantation site planning based on a two-dimensional image and path planning based on a three-dimensional reconstruction result. Based on the two-dimensional image information of the cerebral vessels, an implantation site probability density distribution map is obtained through calculation, the calculation is mainly measured according to the image distance between each pixel in the image and the nearest vessel, the closer the pixel is to the vessel in the two-dimensional image coordinate system, the smaller the probability that the pixel corresponds to the implantation site, the greater the corresponding probability is, the greater the distance between each implantation site and the nearest vessel is, the safer the implantation process is, and the lower the probability of penetrating the cerebral vessels is. After the number of implantation sites and the shortest distance between adjacent sites are artificially formulated, the implantation sites meeting the requirements are selected based on the probability density map of the implantation sites, so that the implantation site planning based on the two-dimensional image is realized. The main calculation formula is as follows:
Based on two-dimensional image information of cerebral vessels, a probability density distribution map M is first calculated, and for each pixel point (u, v) in the distribution map, the probability density value is defined to be proportional to the distance d (u, v) from the nearest vessel, namely
M(u,v)=kd(u,v)
Normalizing the image, wherein the normalized image isNormalization is calculated by the following formula:
After the corresponding normalized two-dimensional probability density distribution map is calculated, N implantation sites are randomly generated in a region with the initial threshold value larger than 0.5, wherein N represents the number of implantation sites set by people, and the corresponding implantation site positions are defined as Meanwhile, the shortest distance between two implantation sites is s, and the implantation site P is optimized based on the conditions:
Finally, obtaining the optimal implantation site planning through optimization iteration
Similarly, based on three-dimensional reconstruction information of cerebral vessels, an implantation path probability density distribution map is obtained through calculation, and the calculation is mainly based on measurement of the spatial distance between each voxel point in a three-dimensional voxel space and the nearest vessel. Combining the completed two-dimensional implantation site planning, taking the two-dimensional implantation site planning as a starting point, and calculating to obtain a three-dimensional implantation path meeting the requirements on the premise of avoiding blood vessels after the implantation depth is artificially formulated, so as to realize the path planning based on the three-dimensional reconstruction result.
In the animal experiment of implanting the brain electrode, the method provided by the invention realizes that each electrode is implanted into a bleeding site smaller than 150 microns, so that the success rate of the operation reaches more than 80%.
The invention also provides a photoacoustic microscopic imaging brain electrode implantation site and a system for planning the path of the photoacoustic microscopic imaging brain electrode implantation site, wherein the system for planning the photoacoustic microscopic imaging brain electrode implantation site and the path of the photoacoustic microscopic imaging brain electrode implantation site can be realized through executing the flow steps of the method for planning the photoacoustic microscopic imaging brain electrode implantation site and the path of the photoacoustic microscopic imaging brain electrode implantation site, namely, a person skilled in the art can understand the method for planning the photoacoustic microscopic imaging brain electrode implantation site and the path of the photoacoustic microscopic imaging brain electrode implantation site as a preferred implementation mode of the system for planning the photoacoustic microscopic imaging brain electrode implantation site and the path of the photoacoustic microscopic imaging brain electrode implantation site.
Specifically, a photoacoustic microscopy imaging brain electrode implantation site and a system for path planning thereof comprise:
The image extraction module is used for extracting an input photoacoustic microscopic scanning imaging result by using a cerebrovascular enhancement extraction algorithm to obtain cerebrovascular information, wherein the cerebrovascular information comprises two-dimensional image information and a three-dimensional reconstruction result of the cerebral blood vessel;
And the implantation site and path acquisition module is used for processing the cerebrovascular information by using a brain electrode implantation site path planning algorithm to obtain an implantation site on a two-dimensional image and an implantation path in a three-dimensional structure.
The image extraction module includes:
The module M1.1 uses the trained denoising model to denoise the acquired photoacoustic microscopic image, and improves the signal to noise ratio;
the trained image enhancement model is used for enhancing the photoacoustic microscopy image, the imaging fuzzy area, the area with the resolution not reaching the preset value and the imaging deformation area are corrected, and the resolution of the photoacoustic microscopy image is improved;
the module M1.3 is used for dividing blood vessels in the photoacoustic microscopy image by using a trained blood vessel division model, removing background and other tissue structure information, and obtaining two-dimensional image information of cerebral blood vessels;
And a module M1.4, combining the depth information of the photoacoustic microscopic imaging with the two-dimensional image information of the cerebral blood vessel, and obtaining a three-dimensional reconstruction result of the cerebral blood vessel through back projection calculation.
The denoising model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting a noise ratio photoacoustic microscopic image for removing noise, training of the deep neural network is based on supervised learning, and the supervision information is a denoising and noise adding process.
The image enhancement model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting the photoacoustic microscopic image with enhanced resolution, training of the deep neural network is based on supervised learning, and supervision information is the recovery difference between the original high-resolution image and the downsampled image.
The blood vessel segmentation model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopic image and outputting a mask image corresponding to a blood vessel region, training of the deep neural network is based on supervised learning, and supervision information is rough segmentation labeling of blood vessels.
The implantation point and path acquisition module comprises:
Selecting starting points for setting the number of implantation sites in a two-dimensional image plane to ensure that the distance between every two implantation sites is more than or equal to the shortest distance between the appointed adjacent sites, optimizing the target to ensure that the sum of the cumulative probability values of all implantation sites is maximum, and optimizing the target to ensure that the distance between every two sites is more than or equal to the shortest distance between the adjacent sites;
and a module M2.2, calculating a three-dimensional cerebrovascular probability density map to acquire an implantation path, selecting an implantation direction with the determined implantation site as a starting point in a three-dimensional space, and optimizing the target to maximize the sum of cumulative probability values on all implantation paths under the condition of designating implantation depth.
Those skilled in the art will appreciate that the invention provides a system and its individual devices, modules, units, etc. that can be implemented entirely by logic programming of method steps, in addition to being implemented as pure computer readable program code, in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Therefore, the system and the devices, modules and units thereof provided by the invention can be regarded as a hardware component, and the devices, modules and units for realizing various functions included in the system can be regarded as structures in the hardware component, and the devices, modules and units for realizing various functions can be regarded as structures in the hardware component as well as software modules for realizing the method.
The foregoing describes specific embodiments of the present application. It is to be understood that the application is not limited to the particular embodiments described above, and that various changes or modifications may be made by those skilled in the art within the scope of the appended claims without affecting the spirit of the application. The embodiments of the application and the features of the embodiments may be combined with each other arbitrarily without conflict.

Claims (7)

1. A method for photoacoustic microscopy imaging brain electrode implantation site and path planning thereof, comprising:
Extracting an input photoacoustic microscopic scanning imaging result by using a cerebral blood vessel enhancement extraction algorithm to obtain cerebral blood vessel information, wherein the cerebral blood vessel information comprises two-dimensional image information and a three-dimensional reconstruction result of cerebral blood vessels;
The method comprises the steps of obtaining implantation sites and paths, namely processing cerebral vessel information by using a cerebral electrode implantation site path planning algorithm to obtain implantation sites on a two-dimensional image and implantation paths in a three-dimensional structure;
the image extraction step includes:
S1.1, denoising the acquired photoacoustic microscopic image by using a trained denoising model, and improving the signal-to-noise ratio;
S1.2, enhancing the photoacoustic microscopy image by using a trained image enhancement model, correcting an imaging fuzzy region, a region with resolution not reaching a preset value and an imaging deformation region, and improving the resolution of the photoacoustic microscopy image;
S1.3, segmenting blood vessels in the photoacoustic microscopy image by using a trained blood vessel segmentation model, and removing background and other tissue structure information to obtain two-dimensional image information of cerebral blood vessels;
Step S1.4, combining the depth information of the photoacoustic microscopic imaging with the two-dimensional image information of the cerebral vessels, and carrying out back projection calculation to obtain a three-dimensional reconstruction result of the cerebral vessels;
the implantation site and path acquisition step includes:
s2.1, calculating a two-dimensional cerebrovascular probability density map to obtain implantation sites, selecting starting points for setting the number of implantation sites in a two-dimensional image plane to ensure that the distance between every two implantation sites is more than or equal to the shortest distance between the appointed adjacent sites, and optimizing the target to ensure that the sum of the cumulative probability values of all implantation sites is maximum, wherein the optimizing condition is that the distance between every two implantation sites is more than or equal to the shortest distance between the adjacent sites;
And S2.2, calculating a three-dimensional brain vessel probability density map to acquire an implantation path, selecting an implantation direction with the determined implantation site as a starting point in a three-dimensional space, and optimizing the target to maximize the sum of cumulative probability values on all implantation paths under the condition of designating implantation depth.
2. The method for planning the implantation site and the path of the brain electrode according to claim 1, wherein the denoising model comprises a deep neural network for inputting an original photoacoustic microscopic image and outputting a noise-removed signal-to-noise ratio photoacoustic microscopic image, wherein training of the deep neural network is based on supervised learning, and the supervision information is a denoising and noise adding process.
3. The method for planning the implantation site and the path of the brain electrode according to claim 1, wherein the image enhancement model comprises a deep neural network for inputting an original photoacoustic microscopic image and outputting a photoacoustic microscopic image with enhanced resolution, training of the deep neural network is based on supervised learning, and the supervision information is a recovery difference between the original high-resolution image and the downsampled image.
4. The photoacoustic microscopy imaging brain electrode implantation site and the path planning method thereof according to claim 1, wherein the blood vessel segmentation model comprises a deep neural network, wherein the deep neural network is used for inputting an original photoacoustic microscopy image and outputting a mask image corresponding to a blood vessel region, training of the deep neural network is based on supervised learning, and the supervision information is rough segmentation labeling of the blood vessel.
5.A system for photoacoustic microscopy imaging of brain electrode implantation sites and path planning thereof, comprising:
The image extraction module is used for extracting an input photoacoustic microscopic scanning imaging result by using a cerebrovascular enhancement extraction algorithm to obtain cerebrovascular information, wherein the cerebrovascular information comprises two-dimensional image information and a three-dimensional reconstruction result of the cerebral blood vessel;
The implantation site and path acquisition module is used for processing the cerebrovascular information by using a brain electrode implantation site path planning algorithm to obtain an implantation site on a two-dimensional image and an implantation path in a three-dimensional structure;
the image extraction module includes:
The module M1.1 uses the trained denoising model to denoise the acquired photoacoustic microscopic image, and improves the signal to noise ratio;
the trained image enhancement model is used for enhancing the photoacoustic microscopy image, the imaging fuzzy area, the area with the resolution not reaching the preset value and the imaging deformation area are corrected, and the resolution of the photoacoustic microscopy image is improved;
the module M1.3 is used for dividing blood vessels in the photoacoustic microscopy image by using a trained blood vessel division model, removing background and other tissue structure information, and obtaining two-dimensional image information of cerebral blood vessels;
The module M1.4 combines the depth information of the photoacoustic microscopic imaging with the two-dimensional image information of the cerebral vessels, and obtains the three-dimensional reconstruction result of the cerebral vessels through back projection calculation;
The implantation site and path acquisition module comprises:
Selecting starting points for setting the number of implantation sites in a two-dimensional image plane to ensure that the distance between every two implantation sites is more than or equal to the shortest distance between the appointed adjacent sites, optimizing the target to ensure that the sum of the cumulative probability values of all implantation sites is maximum, and optimizing the target to ensure that the distance between every two implantation sites is more than or equal to the shortest distance between the adjacent sites;
and a module M2.2, calculating a three-dimensional cerebrovascular probability density map to acquire an implantation path, selecting an implantation direction with the determined implantation site as a starting point in a three-dimensional space, and optimizing the target to maximize the sum of cumulative probability values on all implantation paths under the condition of designating implantation depth.
6. The system for photoacoustic microscopy imaging brain electrode implantation site and path planning thereof according to claim 5, wherein the denoising model comprises a deep neural network for inputting an original photoacoustic microscopy image and outputting a noise-removed signal-to-noise ratio photoacoustic microscopy image, wherein training of the deep neural network is based on supervised learning, and the supervision information is a denoising and noise adding process.
7. The system for photoacoustic microscopy imaging brain electrode implantation site and path planning thereof according to claim 5, wherein the image enhancement model comprises a deep neural network for inputting an original photoacoustic microscopy image and outputting an enhanced resolution photoacoustic microscopy image, wherein training of the deep neural network is based on supervised learning, and the supervision information is a recovery difference between the original high resolution image and the downsampled image.
CN202411761971.8A 2024-12-03 2024-12-03 Method and system for photoacoustic microscopy imaging of brain electrode implantation sites and path planning Active CN119564343B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202411761971.8A CN119564343B (en) 2024-12-03 2024-12-03 Method and system for photoacoustic microscopy imaging of brain electrode implantation sites and path planning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202411761971.8A CN119564343B (en) 2024-12-03 2024-12-03 Method and system for photoacoustic microscopy imaging of brain electrode implantation sites and path planning

Publications (2)

Publication Number Publication Date
CN119564343A CN119564343A (en) 2025-03-07
CN119564343B true CN119564343B (en) 2025-09-30

Family

ID=94810129

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202411761971.8A Active CN119564343B (en) 2024-12-03 2024-12-03 Method and system for photoacoustic microscopy imaging of brain electrode implantation sites and path planning

Country Status (1)

Country Link
CN (1) CN119564343B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN120899390B (en) * 2025-10-09 2026-01-02 北京智冉医疗科技有限公司 Planning method, device and storage medium for electrode implantation point during brain surface pulsation

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022127219A1 (en) * 2020-12-15 2022-06-23 上海微创卜算子医疗科技有限公司 Simulation system and readable storage medium
CN115068082A (en) * 2022-06-20 2022-09-20 中国科学院脑科学与智能技术卓越创新中心 Electrode implantation method and system based on machine vision

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
MY177484A (en) * 2013-03-15 2020-09-16 Synaptive Medical Inc Systems and methods for navigation and simulation of minimally invasive therapy
WO2015166503A1 (en) * 2014-05-01 2015-11-05 Yeda Research And Development Co. Ltd. Multimodal transcranial brain optical imaging
CN116269216A (en) * 2023-02-10 2023-06-23 华南师范大学 Photoacoustic brain imaging system and method based on PVDF-ITO transparent transducer cranium window
CN116459006A (en) * 2023-03-22 2023-07-21 华南师范大学 Photoacoustic Microsurgery Navigation Method and System Based on Synchronous Positioning and Mapping

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022127219A1 (en) * 2020-12-15 2022-06-23 上海微创卜算子医疗科技有限公司 Simulation system and readable storage medium
CN115068082A (en) * 2022-06-20 2022-09-20 中国科学院脑科学与智能技术卓越创新中心 Electrode implantation method and system based on machine vision

Also Published As

Publication number Publication date
CN119564343A (en) 2025-03-07

Similar Documents

Publication Publication Date Title
CN107545584B (en) Method, device and system for positioning region of interest in medical image
US12193784B2 (en) Method and device for target identification, electronic apparatus, storage medium and neuromodulation apparatus
CN107203741B (en) Blood vessel extraction method, device and system
CN113367679B (en) Target point determination method, device, equipment and storage medium
D'Haese et al. Computer-aided placement of deep brain stimulators: from planningto intraoperative guidance
CN105913432B (en) Aorta extracting method and device based on CT sequence images
EP4318393A1 (en) Surgical path processing system, method, apparatus and device, and storage medium
CN113367681B (en) Target point determination method, device, equipment and storage medium
WO2001089386A1 (en) Automatic segmentation of prostate, rectum and urethra in ultrasound imaging
CN110490851A (en) Galactophore image dividing method, apparatus and system based on artificial intelligence
CN119564343B (en) Method and system for photoacoustic microscopy imaging of brain electrode implantation sites and path planning
US12217430B2 (en) Systems and methods for processing x-ray images
CN116912098B (en) Methods, systems, devices, and storage media for medical image processing in interventional surgery
CN114066827B (en) Nucleus region determination method, device, computer equipment and readable storage medium
US20250131566A1 (en) Methods, systems, devices, and storage media for image-assisted interventional surgery
CN115461790A (en) Method and apparatus for classifying structure in image
CN115552479A (en) Method and apparatus for classifying structures in images
CN115552480A (en) Method and apparatus for classifying structures in an image
Ilyasova et al. Biomedical data analysis based on parallel programming technology application for computation features' effectiveness
CN120296569B (en) Auxiliary analysis method, device, equipment and medium for neural circuit damage location
Rashid et al. Analyzing functional magnetic resonance brain images with opencv2
Vadivu et al. Hybrid Mobile-Spinalnet with feature extraction for brain tumor detection using MRI images
Divya et al. Vision Transformer-based Glioma Classification using Multi-Modal MRI and Wavelet Fusion
EP4345747A1 (en) Medical image data processing technique
CN119280713B (en) Tumor motion estimation prediction system and method for radiotherapy robot

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
GR01 Patent grant
GR01 Patent grant