CN117854684A - Intelligent delineating method for segmented prompt type boron neutron capture treatment target area - Google Patents
Intelligent delineating method for segmented prompt type boron neutron capture treatment target area Download PDFInfo
- Publication number
- CN117854684A CN117854684A CN202410060121.9A CN202410060121A CN117854684A CN 117854684 A CN117854684 A CN 117854684A CN 202410060121 A CN202410060121 A CN 202410060121A CN 117854684 A CN117854684 A CN 117854684A
- Authority
- CN
- China
- Prior art keywords
- intelligent
- target area
- delineating
- target
- neutron capture
- 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
- 238000000034 method Methods 0.000 title claims abstract description 54
- ZOXJGFHDIHLPTG-UHFFFAOYSA-N Boron Chemical compound [B] ZOXJGFHDIHLPTG-UHFFFAOYSA-N 0.000 title claims abstract description 38
- 229910052796 boron Inorganic materials 0.000 title claims abstract description 37
- 238000011282 treatment Methods 0.000 title claims abstract description 31
- 238000013528 artificial neural network Methods 0.000 claims abstract description 27
- 230000011218 segmentation Effects 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims abstract description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 37
- 238000002560 therapeutic procedure Methods 0.000 claims description 17
- 238000003708 edge detection Methods 0.000 claims description 13
- 238000010801 machine learning Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 230000001537 neural effect Effects 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 230000000877 morphologic effect Effects 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 238000012805 post-processing Methods 0.000 claims description 2
- 206010028980 Neoplasm Diseases 0.000 abstract description 17
- 201000011510 cancer Diseases 0.000 abstract description 14
- 210000000920 organ at risk Anatomy 0.000 abstract 2
- 238000001959 radiotherapy Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000002708 enhancing effect Effects 0.000 description 4
- 238000003709 image segmentation Methods 0.000 description 4
- 238000002595 magnetic resonance imaging Methods 0.000 description 4
- 150000001875 compounds Chemical class 0.000 description 3
- 210000000056 organ Anatomy 0.000 description 3
- 238000002600 positron emission tomography Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000002591 computed tomography Methods 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000005036 nerve Anatomy 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000002626 targeted therapy Methods 0.000 description 2
- 230000008685 targeting Effects 0.000 description 2
- 230000001225 therapeutic effect Effects 0.000 description 2
- 238000011269 treatment regimen Methods 0.000 description 2
- 206010038111 Recurrent cancer Diseases 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000002512 chemotherapy Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000003702 image correction Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 231100001231 less toxic Toxicity 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06V10/766—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using regression, e.g. by projecting features on hyperplanes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Radiation-Therapy Devices (AREA)
Abstract
The invention discloses an intelligent sketching method of a segmented prompt type boron neutron capture treatment target area, which comprises the following steps: reading in medical image data, converting pixel information into three-bit voxel model information, and extracting and processing image features in medical images; constructing an intelligent target area sketching neural network, wherein the intelligent target area sketching neural network takes a segmentation prompt type method as a basic frame, and utilizes structural information to guide an intelligent sketching process of a boron neutron capture treatment target area; extracting characteristics of a target area from image characteristics in the medical image, and training and optimizing the intelligent delineating neural network of the target area; and (3) carrying out target region delineation on the new medical image by using the trained target region intelligent delineation neural network, and separating the target region from the non-target region. The method can quickly and accurately divide the target tumor and the organ-at-risk area thereof in the boron neutron capture treatment process, and better mark the cancer tissues needing to be mainly killed and the organ-at-risk tissues needing to be mainly protected.
Description
Technical Field
The invention relates to an intelligent target area sketching method, in particular to an intelligent target area sketching method for a segmented prompt type boron neutron capture therapy.
Background
The treatment means for cancer mainly include: surgical treatment, chemotherapy, and radiation therapy. For patients with malignant tumors, about 70% of them require radiation therapy, and about 40% of them can be cured by radiation therapy. Through the development of radiation therapy technology in the last century, scientists continue to explore and develop innovative therapies that are more effective and less toxic to improve local tumor control, patient survival and quality of life. Boron neutron capture therapy (Boron Neutron Capture Therapy, BNCT) is taken as a new generation innovative radiotherapy method, is a binary targeting tumor radiotherapy method combining boron-containing targeting drugs and neutron irradiation, can selectively kill cancer cells with complex shapes without damaging normal tissues, and is one of the most advanced cancer treatment methods internationally at present. Theoretically, it may be an ideal treatment for many types of cancer and has great success in treating advanced, recurrent cancers
Boron neutron capture therapy is a radiation therapy method used to treat specific types of cancer. In BNCT, a physician injects a boron-containing compound into a patient and then subjects the patient to neutron beam irradiation. The neutron beam will react with the boron atoms in the body to produce energetic particles to kill cancer cells. BNCT is a targeted therapy that can selectively destroy cancer cells without affecting healthy tissue. In BNCT, the importance of determining the target volume is very high. Because BNCT is a targeted therapy, only after the exact knowledge of the location and extent of the cancer is achieved, the boron-containing compound can be precisely injected and the neutron beam can be precisely irradiated to the location of the cancer cells. If the target area is not accurately determined, the therapeutic effect may be reduced or even the desired therapeutic effect may not be achieved. In addition, determining the target area may also help the physician to better assess the risk and potential side effects of the treatment and to formulate a more effective treatment regimen. Thus, in BNCT, it is important to accurately determine the target area, and a safer, more effective treatment can be provided to the patient.
In BNCT, the importance of target segmentation is also very high. Target segmentation refers to the further subdivision of the cancer site into smaller areas or parts in order to more accurately determine the injection location of the boron containing compound and the irradiation location of the neutron beam. By target segmentation, the accuracy of treatment can be increased to a higher level, helping to maximize destruction of cancer cells while maximizing protection of surrounding healthy tissue. Furthermore, the target segmentation may also help doctors to better assess the risk and potential side effects of treatment and formulate more effective treatment regimens. Therefore, in BNCT, target segmentation is a very important step, and can improve the accuracy and effect of treatment and provide safer and more effective treatment for patients.
The intelligent target region segmentation refers to analyzing and processing medical image data by utilizing a computer algorithm and an artificial intelligence technology so as to realize more accurate and automatic target region segmentation. In BNCT, the necessity for intelligent segmentation of the target is also very high. Conventional manual segmentation of the target region typically requires a great deal of time and effort from the physician, and is also susceptible to subjective factors, resulting in inaccurate and stable segmentation results. The intelligent segmentation of the target area can help doctors to segment the target area more quickly and accurately, and can reduce the interference of human errors and subjective factors, thereby improving the treatment precision and effect. In addition, the target region intelligent segmentation can also improve the standardization degree of the target region segmentation, so that the treatment result is easier to compare and evaluate. Therefore, the necessity of intelligent segmentation of the target area in BNCT is very high, the treatment effect and safety can be improved, and better treatment experience is provided for patients.
Conventional radiation simulation manikins are mostly composed of simple mathematical geometric models, such as isotropic cubes, cylinders, spheres, etc. As anatomical models continue to optimize, the internal tissue structures of an organism begin to be expressed by simple mathematical models through planar, cylindrical, elliptical, and spherical equations. However, in the radiation simulation calculation, these simple structures tend to bring about deviation of calculation results, and the reliability is also reduced. Conventional automated target delineation methods typically require pretreatment and manual labeling, limiting their accuracy and efficiency in practical applications.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a sectional prompt type boron neutron capture treatment target area intelligent sketching method, after a voxel model of a human body is established based on medical image data, a series of characteristic information existing in the voxel model is utilized to divide an organ area or other interested areas in the model, and target area information files such as RT-Struct files (files for storing medical image information) in a dicom file (Digital Imaging and Communications in Medicine, a type of medical digital imaging and communication standard) are obtained by utilizing the divided information to provide a geometric model required by the Monte Carlo program for completing the related calculation of radiation dose in a simulated human body model.
The technical scheme of the invention is realized as follows:
a sectional prompt type boron neutron capture treatment target area intelligent sketching method comprises the following steps:
s1, reading in medical image data, converting pixel information into three-bit voxel model information, and extracting and processing image features in medical images;
s2, constructing an intelligent target zone delineating neural network, wherein the intelligent target zone delineating neural network takes a segmentation prompt method as a basic framework and comprises a branch structure of a neural differential equation, and the branch structure guides an intelligent delineating process of the boron neutron capture treatment target zone by learning a mapping relation of structural information;
s3, extracting characteristics of a target area from image characteristics in the medical image, taking the characteristics as input of the intelligent target area delineating neural network, classifying or regression predicting the target area by adopting a machine learning algorithm, and training and optimizing the intelligent target area delineating neural network;
and S4, performing target region sketching on the new medical image by using the trained target region intelligent sketching neural network, and separating the target region from the non-target region.
Further, the step S1 includes the steps of: the medical image is pre-processed including, but not limited to, noise removal, filtering, rectification, normalization, segmentation, enhancement, and smoothing.
Further, the S1 medical image data includes, but is not limited to: such as CT images, MRI images, and PET images.
Further, the method for extracting the image features in the medical image in S1 includes, but is not limited to, edge detection and morphological operations.
Further, the intelligent delineating process of the boron neutron capturing treatment target area in the step S2 includes using a rapid edge detection algorithm, an adaptive edge detection algorithm or a rapid region growing algorithm to carry out target area delineating so as to improve the speed and the precision of target area delineating; parallel computation is used to accelerate the computation process of the target volume delineation algorithm.
Further, the characteristics of the target region in S3 include, but are not limited to, shape information, texture information, and gray scale information.
Further, the method of classifying the target region or predicting regression by the machine learning algorithm in S3 includes, but is not limited to, a support vector machine, a random forest, and a neural network.
Further, the step S4 further includes the steps of: post-processing of the sketched results includes, but is not limited to, removing noise and smoothing boundaries.
Further, the algorithm for target region delineation in S4 includes, but is not limited to, a fast edge detection algorithm, an adaptive edge detection algorithm, and a fast region growing algorithm.
Further, the step S4 further includes the steps of: according to different types of medical images, a corresponding model library is established, and the existing annotation data is utilized to train the target area to intelligently sketch the neural network, so that automatic target area sketch is realized.
Compared with the prior art, the method has the beneficial effects that the method can quickly and accurately divide the target tumor and the endangered organ area thereof in the boron neutron capture treatment process, and better mark the cancer tissues needing to be seriously killed and the endangered organ tissues needing to be seriously protected.
Drawings
FIG. 1 is a flow chart of one embodiment of a method for intelligently delineating a segmented cued boron neutron capture therapy target region of the present invention;
fig. 2 is a schematic diagram of a target intelligent delineating neural network in a sectional prompt type boron neutron capture treatment target intelligent delineating method of the 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.
Referring to fig. 1, the present invention aims to provide a method for intelligently delineating a target area for boron neutron capture therapy with a segmented prompt, which comprises the following steps:
s1, reading in medical image data, converting pixel information into three-bit voxel model information, and extracting and processing image features in medical images;
s2, constructing an intelligent target zone delineating neural network, wherein the intelligent target zone delineating neural network takes a segmentation prompt method as a basic framework and comprises a branch structure of a neural differential equation, and the branch structure guides an intelligent delineating process of the boron neutron capture treatment target zone by learning a mapping relation of structural information;
s3, extracting characteristics of a target area from image characteristics in the medical image, taking the characteristics as input of the intelligent target area delineating neural network, classifying or regression predicting the target area by adopting a machine learning algorithm, and training and optimizing the intelligent target area delineating neural network;
and S4, performing target region sketching on the new medical image by using the trained target region intelligent sketching neural network, and separating the target region from the non-target region.
In step S1, medical information including Hounsfield Units (HU) stored in a medical image is input into a computer, and pixel information is converted into three-dimensional voxel model information using a specific algorithm and model. For convenience in post-computer processing, it is generally desirable to pre-process the medical image, including but not limited to removing noise, filtering, enhancing, smoothing, and the like.
Noise and distortion are often present in medical image data, which can negatively impact the accuracy of automatically delineating a target region. Therefore, before automatic delineation, some preprocessing steps are required to be performed on the medical image data to reduce the influence of noise and distortion, thereby improving the accuracy of automatic delineation. Such as: filtering, filtering the medical image by using a proper filter to remove noise and blurring; enhancing, namely enhancing the medical image by using an image enhancement algorithm to enhance the edge and texture information of the target area; correcting deviation, for medical images with distortion, image correction processing such as distortion correction and artifact removal can be performed; before automatic sketching, the medical images need to be standardized into the same gray value range so as to eliminate the difference between different images; the image segmentation algorithm may be used to segment the medical image into different regions prior to automatic delineation to reduce the effects of noise and background, thereby better identifying and delineating the target region.
In step S2, an efficient target region delineating algorithm, such as a fast edge detection algorithm, a self-adaptive edge detection algorithm, a fast region growing algorithm, etc., may be selected, so that the speed and accuracy of target region delineating may be significantly improved. Adjusting algorithm parameters: according to specific application scenes, the speed and accuracy of the algorithm can be further optimized by adjusting parameters of the target region sketching algorithm, such as a threshold value of the edge detection algorithm, growth conditions of the region growing algorithm and the like. Using parallel computing: the calculation process of the target volume delineation algorithm can be accelerated by using parallel calculation technology, such as GPU (Graphic Processing Unit, graphic processor) acceleration, distributed calculation and the like, so that the speed and the accuracy are improved.
The target area intelligent delineating neural network structure is shown in fig. 2, the target area intelligent delineating neural network is characterized in that a subsection prompt type method is used as a basic framework for boron neutron capture treatment target area delineating, a branch structure based on a nerve differential equation is designed in the subsection prompt type method, the process of intelligent delineating of the boron neutron capture treatment target area is guided by utilizing structural information through the structural information mapping relation in the branch learning production process, the network residual error linking structure is improved based on the nerve differential equation, the adaptability of a characteristic network to diversified characteristics is enhanced, and finally the accuracy of the intelligent delineating of the boron neutron capture treatment target area is improved, so that a few tumor areas are prevented from being forgotten and insufficient dose distribution is avoided.
In the process of target region sketching, machine learning and image segmentation technology can be combined to realize automatic sketching of target region. Specifically, the following steps may be employed: data preprocessing: firstly, medical images need to be preprocessed, such as filtering, denoising, enhancing and the like, so as to improve the image quality and reduce interference; feature extraction: next, it is necessary to extract the characteristics of the target region, such as shape, texture, and gray scale, from the medical image as input to the machine learning model; model training: then, a machine learning algorithm (such as a support vector machine, a random forest or a neural network) is adopted to classify or predict regression of the target area, and the model is trained and optimized; image segmentation: dividing the new medical image by using the trained model, and separating the target area from the non-target area; post-treatment: finally, the segmentation results may be post-processed, such as to remove noise and smooth boundaries, to obtain more accurate and stable target volume delineation results.
To improve the application of the automatic target region delineation technology in different types of medical images, more powerful and rich model training is required, and in order to accelerate the image segmentation speed and adapt to different image types, the information data types contained in the different medical images need to be known firstly, and then suitable target region delineation algorithms and parameters are selected for different types of medical images (such as CT (Computed Tomography, electronic computer tomography) images, MRI (magnetic resonance imaging Magnetic resonance imaging) images, PET (Positron Emission Tomography ) images and the like) so as to adapt to the characteristics of different image types.
Taking into account the differences between different medical images and considering the image quality, for example, the differences may exist in the quality of different medical images, such as noise, blurring and the like, a corresponding preprocessing method needs to be adopted to improve the accuracy and stability of the target region delineation. The information features can be used for establishing a corresponding model library, such as a typical head-neck model, a chest-body model and the like, aiming at different types of medical images, the corresponding model library can be established, and the model is trained by using the existing labeling data so as to realize automatic target region sketching.
Because a part of Monte Carlo post-processing program of the target area automatic sketching system needs to have a sketching sequence with certain requirements, the method and the device are used for building by combining a corresponding model library, and simultaneously, a dicom file sequence which can meet the geometric processing of an MC program can be generated so as to meet the requirements of the post-processing program on generating corresponding geometry and a corresponding voxel grid model. While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that changes and modifications may be made without departing from the principles of the invention, such changes and modifications are also intended to be within the scope of the invention.
Claims (10)
1. A method for intelligently sketching a segmented prompt type boron neutron capture treatment target area is characterized by comprising the following steps:
s1, reading in medical image data, converting pixel information into three-bit voxel model information, and extracting and processing image features in medical images;
s2, constructing an intelligent target zone delineating neural network, wherein the intelligent target zone delineating neural network takes a segmentation prompt method as a basic framework and comprises a branch structure of a neural differential equation, and the branch structure guides an intelligent delineating process of the boron neutron capture treatment target zone by learning a mapping relation of structural information;
s3, extracting characteristics of a target area from image characteristics in the medical image, taking the characteristics as input of the intelligent target area delineating neural network, classifying or regression predicting the target area by adopting a machine learning algorithm, and training and optimizing the intelligent target area delineating neural network;
and S4, performing target region sketching on the new medical image by using the trained target region intelligent sketching neural network, and separating the target region from the non-target region.
2. The method for intelligently delineating a segmented cued boron neutron capture therapy target region according to claim 1, wherein the step S1 comprises the steps of: the medical image is pre-processed including, but not limited to, noise removal, filtering, rectification, normalization, segmentation, enhancement, and smoothing.
3. The method for intelligently delineating a segmented cued boron neutron capture therapy target region according to claim 1, wherein the S1 medical image data includes, but is not limited to: such as CT images, MRI images, and PET images.
4. The method for intelligently delineating a segmented cued boron neutron capture therapy target region according to claim 1, wherein the method for extracting image features in the medical image in S1 includes but is not limited to edge detection and morphological operations.
5. The intelligent delineation method of the segmented cueing type boron neutron capture therapy target area according to claim 1, wherein the intelligent delineation process of the boron neutron capture therapy target area in the step S2 comprises the steps of using a rapid edge detection algorithm, an adaptive edge detection algorithm or a rapid region growing algorithm to carry out target area delineation so as to improve the speed and the accuracy of target area delineation; parallel computation is used to accelerate the computation process of the target volume delineation algorithm.
6. The method for intelligently delineating a segmented cued boron neutron capture therapy target according to claim 1, wherein the characteristics of the target in S3 include, but are not limited to, shape information, texture information, and gray scale information.
7. The method for intelligently delineating a target region for segmented cueing type boron neutron capture therapy according to claim 1, wherein the method for classifying or regression predicting the target region by the machine learning algorithm in S3 comprises but is not limited to a support vector machine, a random forest and a neural network.
8. The method for intelligently delineating a segmented cued boron neutron capture therapy target region according to claim 1, wherein S4 further comprises the steps of: post-processing of the sketched results includes, but is not limited to, removing noise and smoothing boundaries.
9. The method for intelligently delineating a target region for segmented cueing type boron neutron capture therapy according to claim 1, wherein the algorithm for delineating the target region in S4 includes but is not limited to a fast edge detection algorithm, an adaptive edge detection algorithm and a fast region growing algorithm.
10. The method for intelligently delineating a segmented cued boron neutron capture therapy target region according to claim 1, wherein S4 further comprises the steps of: according to different types of medical images, a corresponding model library is established, and the existing annotation data is utilized to train the target area to intelligently sketch the neural network, so that automatic target area sketch is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410060121.9A CN117854684A (en) | 2024-01-16 | 2024-01-16 | Intelligent delineating method for segmented prompt type boron neutron capture treatment target area |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410060121.9A CN117854684A (en) | 2024-01-16 | 2024-01-16 | Intelligent delineating method for segmented prompt type boron neutron capture treatment target area |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117854684A true CN117854684A (en) | 2024-04-09 |
Family
ID=90530358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410060121.9A Pending CN117854684A (en) | 2024-01-16 | 2024-01-16 | Intelligent delineating method for segmented prompt type boron neutron capture treatment target area |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117854684A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118211422A (en) * | 2024-04-12 | 2024-06-18 | 华硼中子科技(杭州)有限公司 | Non-uniform dynamic boron concentration distribution model acquisition method and device, storage medium, terminal and computer program product |
-
2024
- 2024-01-16 CN CN202410060121.9A patent/CN117854684A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118211422A (en) * | 2024-04-12 | 2024-06-18 | 华硼中子科技(杭州)有限公司 | Non-uniform dynamic boron concentration distribution model acquisition method and device, storage medium, terminal and computer program product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11386557B2 (en) | Systems and methods for segmentation of intra-patient medical images | |
| CN108770373B (en) | pseudo-CT generation from MR data using a feature regression model | |
| CN108778416B (en) | Systems, methods, and media for pseudo-CT generation from MR data using tissue parameter estimation | |
| CN108815721B (en) | Irradiation dose determination method and system | |
| US10149987B2 (en) | Method and system for generating synthetic electron density information for dose calculations based on MRI | |
| EP3695882A1 (en) | Computing radiotherapy dose distribution | |
| KR102504022B1 (en) | Device for planning a non-invaseve treatment based on artificial intelligence using ct image generated from brain mri image | |
| CN102184334A (en) | Retrieval-based radiotherapy planning system and retrieval method | |
| Xu et al. | An algorithm for efficient metal artifact reductions in permanent seed implants | |
| CN117854684A (en) | Intelligent delineating method for segmented prompt type boron neutron capture treatment target area | |
| CN104636618A (en) | Radiotherapy treatment planning system | |
| CN113129327B (en) | Method and system for generating internal general target area based on neural network model | |
| CN117427286B (en) | Tumor radiotherapy target area identification method, system and equipment based on energy spectrum CT | |
| CN114558251A (en) | Automatic positioning method and device based on deep learning and radiotherapy equipment | |
| Chandran et al. | MemU-Net: A new volumetric dose prediction model using deep learning techniques in radiation treatment planning | |
| Tang et al. | A feasibility study of treatment verification using EPID cine images for hypofractionated lung radiotherapy | |
| Spezialetti et al. | Using deep learning for fast dose refinement in proton therapy | |
| Zhou et al. | Deep-learning Segmentation of Small Volumes in CT images for Radiotherapy Treatment Planning | |
| CN118366614B (en) | Boron neutron capture treatment plan curative effect prediction model construction method and device | |
| Hien et al. | A framework for 3D radiotherapy dose prediction using the deep learning approach. | |
| Arjmandi et al. | Deep learning-based automated liver contouring using a small sample of radiotherapy planning computed tomography images | |
| Miandoab et al. | Extraction of respiratory signal based on image clustering and intensity parameters at radiotherapy with external beam: A comparative study | |
| Gallery et al. | Learning Multi-Catheter Reconstructions for Interstitial Breast Brachytherapy | |
| Schildkraut et al. | Level‐set segmentation of pulmonary nodules in megavolt electronic portal images using a CT prior | |
| JAMPA-NGERN | Study on Quick Prediction of Dose Volume Statistics in Proton Beam Therapy using Deep Learning |
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 |