CN111862021B - Deep learning-based automatic head and neck lymph node and drainage area delineation method - Google Patents
Deep learning-based automatic head and neck lymph node and drainage area delineation method Download PDFInfo
- Publication number
- CN111862021B CN111862021B CN202010670160.2A CN202010670160A CN111862021B CN 111862021 B CN111862021 B CN 111862021B CN 202010670160 A CN202010670160 A CN 202010670160A CN 111862021 B CN111862021 B CN 111862021B
- Authority
- CN
- China
- Prior art keywords
- lymph
- drainage area
- deep learning
- image
- head
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
Abstract
The embodiment of the invention provides a deep learning-based automatic head and neck lymph node and drainage area delineation method, which utilizes the symmetry of a human body structure to divide a head area into a left part and a right part for training and prediction, thereby indirectly increasing the data volume of model training and simultaneously reducing the scale of a deep learning model. The invention uses a delineation strategy of stepwise processing optimization, firstly adopts a deep learning model to realize a lymph drainage area which is easy to be segmented, and then adopts a multi-task deep learning model to optimize the lymph drainage area and segment lymph nodes, thereby fully utilizing the relevance between the lymph drainage area and the lymph nodes and improving the segmentation accuracy of the lymph drainage area and the lymph nodes. The invention implements an Artificial Intelligence (AI) assisted contour delineation method in the radiation therapy planning work flow, can effectively improve the delineation consistency of the work efficiency of medical workers, and improves the radiation therapy accuracy of head and neck tumors.
Description
Technical Field
The invention relates to the field of medical images, in particular to a deep learning-based automatic head and neck lymph node and drainage area delineation method.
Background
Due to the fact that the lymphatic network is widely distributed under the mucosa of the head and the neck, regional lymph node metastasis risk of head and neck tumors is high, and the probability of occult metastasis is as high as 30%. In clinical radiotherapy of patients with head and neck tumors, in addition to high dose irradiation of tumor primary foci (GTV), prophylactic irradiation of lymph nodes (GTVn) and lymph drainage areas (CTVn) with metastasis around the tumor is required. At present, radiotherapy is one of important treatment modes for head and neck tumors, the head and neck have a plurality of important and fine structures, and the precise radiotherapy technology has important clinical significance on the postoperative life quality of patients and is the urgent clinical demand at present. Therefore, the contour delineation in the radiation therapy is the basic guarantee of accurate radiation therapy, the accuracy of the contour delineation directly determines the reliability of radiation therapy dose distribution, and the wrong delineation may bring serious radiation therapy accidents, endangering the life safety of patients. Lymph drainage areas have better contrast and boundary on CT, but lymph nodes are difficult to distinguish on positioning CT or enhanced CT of radiotherapy, and are particularly easy to be confused with adjacent muscles and blood vessels, so that false delineation or missed delineation is caused.
At present, the boundaries of the metastatic lymph nodes and the lymph drainage area are clinically drawn by doctors manually or image segmentation methods based on deep learning, wherein the manual drawing has low efficiency and poor repeatability and depends heavily on the experience level of the drawing doctors and the drawing standard reference. In addition, the lymph node metastasis risk of different types of tumors and tumors in different positions of the head and the neck is greatly different, and the clinical lymph node metastasis delineation is more challenging.
The conventional image segmentation method based on deep learning depends on image labeling and training of a large amount of data, and clinically acquired data usually has the problems of incomplete labeling and non-uniform labeling schemes, so that the amount of practically available training data is very small. Therefore, the existing deep learning-based technology cannot support automatic delineation of low-contrast and irregularly-distributed lymph nodes. In addition, the head and neck lymph drainage area has a large image coverage range, and the delineation depends on image information between layers, so that the training of a segmentation model under the input of a large three-dimensional image cannot be supported under the general hardware configuration.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
Disclosure of Invention
Aiming at the problems in the prior art, the embodiment of the invention provides a method for automatically delineating a head and neck lymph node and a drainage area based on deep learning.
The embodiment of the invention provides a deep learning-based automatic head and neck lymph node and drainage area delineating method, which comprises the following steps:
step (1): retrospectively collecting planned CT image data of a head and neck tumor patient before radiotherapy, and preprocessing the planned CT image data;
step (2): extracting delineation contours of a lymph drainage area and a lymph node which are delineated on a planned CT image of a head and neck tumor patient collected retrospectively in clinic, assigning a region inside the contour line to be 1 and assigning a region outside the contour line to be 0, and obtaining a binary mask image of each organ at risk;
and (3): identifying the head region of a patient from a head neck tumor patient plan CT image to obtain a binary mask image of the head; calculating the center line of the head region along the left and right directions as left and right head region left and right dividing lines, and dividing the plan CT image and the binary mask image of the lymph drainage region and the lymph node left and right;
and (4): three-dimensionally clipping the divided plan CT image and the binary mask image of the lymph drainage area and the lymph node, and extracting regions of interest in the left-right, front-back and head-foot directions based on the delineation result of the lymph drainage area;
and (5): carrying out five-fold cross training and verification on the deep learning model A by using the extracted normalized planned CT image in the region of interest and the binary mask image of the lymph drainage area to obtain a trained lymph drainage area segmentation model A and a delineation result of lymph drainage area verification of all training cases;
and (6): taking the segmentation result of the lymph drainage area in the step (5) and the planned CT image in the region of interest as the input of a deep learning model B, taking the binary mask images of the lymph drainage area and the lymph node as the output of the deep learning model B, and training the multitask deep learning model B to obtain a trained lymph node and lymph drainage area segmentation optimization model B;
and (7): acquiring planned CT image data of a new head and neck tumor patient, preprocessing the planned CT image data according to the step (1-4) to obtain planned CT images in the interested areas on the left side and the right side, and respectively inputting the planned CT images into the trained lymph drainage area segmentation model A to obtain a primary segmentation result of the left lymph drainage area and the right lymph drainage area;
and (8): inputting the planned CT images in the interested areas on the left side and the right side obtained in the step (7) and the preliminary segmentation results of the left lymph drainage area and the right lymph drainage area into a trained lymph node and lymph drainage area segmentation optimization model B to obtain the segmentation results of the lymph nodes on the left side and the right side and the lymph drainage area.
Further, the planning CT image data before radiotherapy of the head and neck tumor patient is preprocessed in the step (1) as follows: the image CT values are truncated to the range of-150,250 and normalized to the range of-1, 1.
Furthermore, in the step (3), a threshold segmentation method is adopted to identify the head region of the patient from the planned CT image of the patient with the head and neck tumor, and the closed region with the CT value larger than-200 is taken as the head region to obtain a binary mask image of the head.
Further, in the step (4), the size of the region of interest in the left-right, front-back and head-foot directions is extracted to be 128 × 192 × 96 based on the delineation result of the lymph drainage region, and the unit of each direction is pixel.
Further, the deep learning model a adopted in the step (5) adopts a five-fold cross training and verification method, which specifically comprises the following steps:
step (a): dividing the extracted planning CT image and the extracted binary mask image of the lymph drainage area in the region of interest into five groups by taking a patient as a unit;
step (b): establishing a deep learning model A, inputting planned CT images in four groups of interested areas in the five groups of data divided in the step (a) as the deep learning model A, and training the deep learning model A-1 by taking a binary mask image of a corresponding lymph drainage area as output to obtain a trained deep learning model A-1;
step (c): taking the planned CT image in the interested area of the remaining group of data as the input of a deep learning model A-1 to obtain the primary segmentation result of the lymph drainage area;
step (d): replacing the verification data, repeating the steps (b) and (c), and respectively training the deep learning models A-2, A-3, A-4 and A-5 to obtain the sketching result of the lymph drainage area corresponding to the verification data.
Further, the training of the deep learning model in the step (6) is performed in two stages, specifically including the following steps:
step i: constructing a deep learning model B, wherein the number of input channels is 2, the first input channel is a planned CT image of an interested area, and the second channel is a preliminary lymph drainage area segmentation result; the number of output channels is 2, the first output channel is the segmentation result of the lymph nodes, and the second output channel is the segmentation result of the lymph drainage area;
step ii: setting a second input channel of the model B as all-zero data, training the deep learning model B, extracting the characteristics of the CT image, segmenting lymph nodes and lymph drainage areas, and training until the model is stable;
step iii: (iii) setting the second channel of model B as the preliminary segmentation result of lymphatic drainage region, and retraining the stable model trained in step (ii) until the stable model is stable.
The method for automatically delineating the head and neck lymph nodes and the drainage area based on deep learning provided by the embodiment of the invention has the following advantages:
1. the invention utilizes the symmetry of the human body structure to divide the head area into a left part and a right part for training and predicting, thereby indirectly increasing the data volume of model training and simultaneously reducing the scale of a deep learning model.
2. The invention uses a delineation strategy of stepwise processing optimization, firstly adopts a deep learning model to realize a lymph drainage area which is easy to be segmented, and then adopts a multi-task deep learning model to optimize the lymph drainage area and segment lymph nodes, thereby fully utilizing the relevance between the lymph drainage area and the lymph nodes and improving the segmentation accuracy of the lymph drainage area and the lymph nodes.
3. The invention implements the AI-assisted contour delineation method in the radiotherapy plan work flow, can effectively improve the delineation consistency of the work efficiency of medical workers, and improves the precision of the radiotherapy of the head and neck tumors.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a flowchart of a deep learning-based method for automatically delineating a head and neck lymph node and a drainage area according to an embodiment of the present invention;
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Throughout the specification and claims, unless explicitly stated otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or component but not the exclusion of any other element or component.
Fig. 1 is a flowchart of an automatic head and neck lymph node and drainage area delineation method based on deep learning according to an embodiment of the present invention, and as shown in fig. 1, the automatic head and neck lymph node and drainage area delineation method based on deep learning according to the present invention includes the following steps:
step (1): the planned CT image data before radiotherapy of head and neck tumor patients (such as nasopharyngeal carcinoma, oropharyngeal cancer, parotid gland cancer, etc.) is collected retrospectively and preprocessed as follows: the CT value of the image is intercepted in the range of-150,250, and then is normalized to the range of-1, so that the image data of each patient is more uniform;
step (2): extracting delineation outlines of a lymph drainage area and a lymph node which are delineated on a planned CT image of a patient with head and neck tumor collected retrospectively in clinic, assigning a region inside the outline to be 1 and a region outside the outline to be 0, and obtaining a binary mask image of each organ at risk;
and (3): and identifying the head region of the patient from the planned CT image of the head-neck tumor patient by adopting a threshold segmentation method, and taking the closed region with the CT value of more than-200 as the head region to obtain a binary mask image of the head. Because the head and neck tumor patients need to adopt a mask to fix the head and neck clinically, the body position is fixed, the central line of the head area along the left and right directions can be calculated and used as a left and right dividing line of the left and right head areas, and the left and right division is carried out on the planning CT image and the binary mask image of the lymph drainage area and the lymph node;
and (4): and performing three-dimensional clipping on the divided planning CT image and the binary mask image of the lymph drainage area and the lymph node. Considering that the resolution of a conventional CT image is about 1mm by 3mm generally, regions of interest with the sizes of the left side, the right side, the front side, the back side and the head and foot directions of 128 by 192 by 96 are extracted based on the delineation result of a lymph drainage region, and the unit of each direction is pixel;
and (5): carrying out five-fold cross training and verification on the deep learning model A by using the extracted normalized planned CT image (input) in the region of interest and the binary mask image (output) of the lymph drainage area to obtain a trained lymph drainage area segmentation model A and a delineation result of lymph drainage area verification of all training cases;
and (4) adopting a five-fold cross training and verification method for the deep learning model A adopted in the step (5). The method specifically comprises the following steps:
step (a): dividing the extracted planning CT image and the extracted binary mask image of the lymph drainage area in the region of interest into five groups by taking a patient as a unit;
step (b): establishing a deep learning model A, inputting planned CT images in four groups of interested areas in the five groups of data divided in the step (a) as the deep learning model A, and training the model A-1 by taking a binary mask image of a corresponding lymph drainage area as output to obtain a trained deep learning model A-1;
a step (c): taking the planned CT image in the region of interest of the remaining group of data as the input of the model A-1 to obtain the primary segmentation result of the lymph drainage area;
a step (d): replacing the verification data, repeating the steps (b) and (c), and respectively training the models A-2, A-3, A-4 and A-5 to obtain the sketching result of the lymph drainage area corresponding to the verification data.
And (6): taking the segmentation result of the lymph drainage area in the step (5) and the planned CT image in the region of interest as the input of a deep learning model B, taking the binary mask images of the lymph drainage area and the lymph node as the output of the deep learning model B, and training the multitask deep learning model B to obtain a trained lymph node and lymph drainage area segmentation optimization model B;
the training of the deep learning model in the step (6) is performed in two stages, and the method specifically comprises the following steps:
step i: and constructing a deep learning model B, wherein the number of input channels is 2, the first input channel is a planned CT image of the region of interest, and the second channel is a preliminary lymph drainage area segmentation result. The number of output channels is 2, the first output channel is the segmentation result of the lymph nodes, and the second output channel is the segmentation result of the lymph drainage area;
step ii: setting a second input channel of the model B as all-zero data, training the deep learning model B, extracting the characteristics of the CT image, segmenting lymph nodes and lymph drainage areas, and training until the model is stable;
step iii: setting the second channel of the model B as the primary segmentation result of the lymph drainage area, and retraining the stable model trained in the step (ii) until the stable model is stable.
And (7): acquiring planned CT image data of a new head and neck tumor patient, preprocessing the planned CT image data according to the step (1-4) to obtain planned CT images in the interested areas on the left side and the right side, and respectively inputting the planned CT images into the trained lymph drainage area segmentation model A to obtain a primary segmentation result of the left lymph drainage area and the right lymph drainage area;
and (8): inputting the planned CT images in the interested areas on the left side and the right side obtained in the step (7) and the preliminary segmentation results of the left lymph drainage area and the right lymph drainage area into a trained lymph node and lymph drainage area segmentation optimization model B to obtain the segmentation results of the lymph nodes on the left side and the right side and the lymph drainage area.
In summary, compared with the prior art, the method provided by the invention utilizes the symmetry of the human body structure to divide the head region into the left part and the right part for training and prediction, thereby indirectly increasing the data volume of model training and simultaneously reducing the scale of the deep learning model. The invention uses a delineation strategy of stepwise processing optimization, firstly adopts a deep learning model to realize a lymph drainage area which is easy to be segmented, and then adopts a multi-task deep learning model to optimize the lymph drainage area and segment lymph nodes, thereby fully utilizing the relevance between the lymph drainage area and the lymph nodes and improving the segmentation accuracy of the lymph drainage area and the lymph nodes. The invention implements the AI-assisted contour delineation method in the radiotherapy plan work flow, can effectively improve the delineation consistency of the work efficiency of medical workers, and improves the precision of the radiotherapy of the head and neck tumors.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on the understanding, the above technical solutions substantially or otherwise contributing to the prior art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (4)
1. A method for automatically delineating a head-neck lymph node and a drainage area based on deep learning is characterized by comprising the following steps:
step (1): retrospectively collecting planned CT image data of a patient with head and neck tumor before radiotherapy, and preprocessing the planned CT image data;
step (2): extracting delineation outlines of a lymph drainage area and a lymph node which are delineated on a planned CT image of a patient with head and neck tumor collected retrospectively in clinic, assigning a region inside the outline to be 1 and a region outside the outline to be 0, and obtaining a binary mask image of each organ at risk;
and (3): identifying the head region of a patient from a head neck tumor patient plan CT image to obtain a binary mask image of the head; calculating the center line of the head region along the left and right directions as left and right head region left and right dividing lines, and dividing the plan CT image and the binary mask image of the lymph drainage region and the lymph node left and right;
and (4): performing three-dimensional cutting on the divided plan CT image and the binary mask image of the lymph drainage area and the lymph node, and extracting the interested areas in the left-right, front-back and head-foot directions based on the sketching result of the lymph drainage area;
and (5): carrying out five-fold cross training and verification on the deep learning model A by using the extracted normalized planned CT image in the region of interest and the binary mask image of the lymph drainage area to obtain a trained lymph drainage area segmentation model A and a delineation result of lymph drainage area verification of all training cases; the deep learning model A adopted in the step (5) adopts a five-fold cross training and verification method, which specifically comprises the following steps:
step (a): dividing the extracted planning CT image and the extracted binary mask image of the lymph drainage area in the region of interest into five groups by taking a patient as a unit;
step (b): establishing a deep learning model A, inputting planned CT images in four groups of interested areas in the five groups of data divided in the step (a) as the deep learning model A, and training the deep learning model A-1 by taking a binary mask image of a corresponding lymph drainage area as output to obtain a trained deep learning model A-1;
step (c): taking the planned CT image in the interested area of the remaining group of data as the input of a deep learning model A-1 to obtain the primary segmentation result of the lymph drainage area;
a step (d): replacing verification data, repeating the steps (b) and (c), and respectively training deep learning models A-2, A-3, A-4 and A-5 to obtain a delineation result of the lymph drainage area corresponding to the verification data;
and (6): taking the delineation result of the lymph drainage area in the step (5) and the planned CT image in the region of interest as the input of a deep learning model B, taking the binary mask images of the lymph drainage area and the lymph node as the output of the deep learning model B, and training the multitask deep learning model B to obtain a trained lymph node and lymph drainage area segmentation optimization model B; the training of the deep learning model in the step (6) is performed in two stages, and specifically comprises the following steps:
step i: constructing a deep learning model B, wherein the number of input channels is 2, the first input channel is a planned CT image of an interested area, and the second input channel is a preliminary lymph drainage area segmentation result; the number of output channels is 2, the first output channel is the segmentation result of lymph nodes, and the second output channel is the segmentation result of lymph drainage areas;
step ii: setting a second input channel of the model B as all-zero data, training the deep learning model B, extracting the characteristics of the CT image, segmenting lymph nodes and lymph drainage areas, and training until the model is stable;
step iii: setting a second input channel of the model B as a preliminary lymph drainage area segmentation result, and retraining the trained stable model in the step (ii) until the model is stable;
and (7): acquiring planned CT image data of a new head and neck tumor patient, preprocessing the planned CT image data according to the steps (1) to (4) to obtain planned CT images in the interested areas on the left side and the right side, and respectively inputting the planned CT images into the trained lymph drainage area segmentation model A to obtain a primary segmentation result of the left lymph drainage area and the right lymph drainage area;
and (8): inputting the planned CT images in the interested areas on the left side and the right side obtained in the step (7) and the preliminary segmentation results of the left lymph drainage area and the right lymph drainage area into a trained lymph node and lymph drainage area segmentation optimization model B to obtain the segmentation results of the lymph nodes on the left side and the right side and the lymph drainage area.
2. The method for automatically delineating a cervical lymph node and drainage area based on deep learning of claim 1, wherein the planned CT image data of a patient with head and neck tumor before radiotherapy in step (1) is pre-processed as follows: the image CT values are truncated to the range of-150,250 and normalized to the range of-1, 1.
3. The method for automatically delineating a cervical lymph node and a drainage region based on deep learning according to claim 1, wherein a threshold segmentation method is adopted in the step (3) to identify the head region of the patient from the planned CT image of the patient with cervical tumor, and a closed region with a CT value greater than-200 is taken as the head region to obtain a binary mask image of the head.
4. The method for automatically delineating a cervical lymph node and drainage area based on deep learning according to claim 1, wherein the size of the region of interest extracted in the left-right, front-back and head-foot directions based on the delineation result of the lymph drainage area in step (4) is 128 × 192 × 96, and the unit of each direction is pixel.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010670160.2A CN111862021B (en) | 2020-07-13 | 2020-07-13 | Deep learning-based automatic head and neck lymph node and drainage area delineation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010670160.2A CN111862021B (en) | 2020-07-13 | 2020-07-13 | Deep learning-based automatic head and neck lymph node and drainage area delineation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111862021A CN111862021A (en) | 2020-10-30 |
CN111862021B true CN111862021B (en) | 2022-06-24 |
Family
ID=72982928
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010670160.2A Active CN111862021B (en) | 2020-07-13 | 2020-07-13 | Deep learning-based automatic head and neck lymph node and drainage area delineation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111862021B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111986213B (en) * | 2020-08-21 | 2022-12-23 | 四川大学华西医院 | Processing method, training method and device of slice image and storage medium |
CN112790782B (en) * | 2021-02-02 | 2022-06-24 | 广州柏视医疗科技有限公司 | Automatic pelvic tumor CTV (computer-to-volume) delineation system based on deep learning |
CN113012144A (en) * | 2021-04-08 | 2021-06-22 | 湘南学院附属医院 | Automatic delineation method and system for lung tumor, computing device and storage medium |
CN113488146B (en) * | 2021-07-29 | 2022-04-01 | 广州柏视医疗科技有限公司 | Automatic delineation method for drainage area and metastatic lymph node of head and neck nasopharyngeal carcinoma |
CN115409739B (en) * | 2022-10-31 | 2023-01-24 | 中山大学肿瘤防治中心(中山大学附属肿瘤医院、中山大学肿瘤研究所) | Method and system for automatically sketching organs at risk |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104524697A (en) * | 2014-12-24 | 2015-04-22 | 昆明市延安医院 | Image-guided nasopharyngeal carcinoma intensity modulated radiation therapy position error method |
CN110705565A (en) * | 2019-09-09 | 2020-01-17 | 西安电子科技大学 | Lymph node tumor region identification method and device |
CN111105424A (en) * | 2019-12-19 | 2020-05-05 | 广州柏视医疗科技有限公司 | Lymph node automatic delineation method and device |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102048550B (en) * | 2009-11-02 | 2013-07-17 | 上海交通大学医学院附属仁济医院 | Method for automatically generating liver 3D (three-dimensional) image and accurately positioning liver vascular domination region |
CN108257134B (en) * | 2017-12-21 | 2022-08-23 | 深圳大学 | Nasopharyngeal carcinoma focus automatic segmentation method and system based on deep learning |
-
2020
- 2020-07-13 CN CN202010670160.2A patent/CN111862021B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104524697A (en) * | 2014-12-24 | 2015-04-22 | 昆明市延安医院 | Image-guided nasopharyngeal carcinoma intensity modulated radiation therapy position error method |
CN110705565A (en) * | 2019-09-09 | 2020-01-17 | 西安电子科技大学 | Lymph node tumor region identification method and device |
CN111105424A (en) * | 2019-12-19 | 2020-05-05 | 广州柏视医疗科技有限公司 | Lymph node automatic delineation method and device |
Non-Patent Citations (2)
Title |
---|
Coarse-to-Fine Stacked Fully Convolutional Nets for Lymph Node Segmentation in Ultrasound Images;Yizhe Zhang et al.;《2016 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)》;20170119;第443-448页 * |
鼻咽癌调强放疗中对危及器官勾画及计划优化改进的剂量学探讨;徐林 等;《中山大学学报(医学科学版)》;20150930;第36卷(第5期);第745-752页 * |
Also Published As
Publication number | Publication date |
---|---|
CN111862021A (en) | 2020-10-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111862021B (en) | Deep learning-based automatic head and neck lymph node and drainage area delineation method | |
US10769791B2 (en) | Systems and methods for cross-modality image segmentation | |
US11455732B2 (en) | Knowledge-based automatic image segmentation | |
US11100647B2 (en) | 3-D convolutional neural networks for organ segmentation in medical images for radiotherapy planning | |
Egger et al. | Pituitary adenoma volumetry with 3D Slicer | |
CN107077736B (en) | System and method for segmenting medical images based on anatomical landmark-based features | |
WO2023221954A1 (en) | Pancreatic tumor image segmentation method and system based on reinforcement learning and attention | |
Mahdavi et al. | Semi-automatic segmentation for prostate interventions | |
CN111028914B (en) | Artificial intelligence guided dose prediction method and system | |
CN112419338B (en) | Head and neck endangered organ segmentation method based on anatomical prior knowledge | |
CN111862022B (en) | Automatic delineation method for organs at risk in whole body multi-part radiotherapy | |
US9098912B2 (en) | Method, system and computer readable medium for automatic segmentation of a medical image | |
CN109636806B (en) | Three-dimensional nuclear magnetic resonance pancreas image segmentation method based on multi-step learning | |
US9727975B2 (en) | Knowledge-based automatic image segmentation | |
CN111738989A (en) | Organ delineation method and device | |
Lei et al. | Male pelvic multi‐organ segmentation on transrectal ultrasound using anchor‐free mask CNN | |
Garg et al. | A survey of prostate segmentation techniques in different imaging modalities | |
Luximon et al. | Machine‐assisted interpolation algorithm for semi‐automated segmentation of highly deformable organs | |
Ananth et al. | Graph Cutting Tumor Images | |
CN114581474A (en) | Automatic clinical target area delineation method based on cervical cancer CT image | |
Wang et al. | Multi-view fusion segmentation for brain glioma on CT images | |
Zhang et al. | Automatic parotid gland segmentation in MVCT using deep convolutional neural networks | |
CN117244181A (en) | Dose analysis method and device based on radiotherapy risk organ outline | |
CN110120052A (en) | A kind of target area image segmenting system and device | |
CN115187577B (en) | Automatic drawing method and system for breast cancer clinical target area based on 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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20211129 Address after: 510530 room 306, phase I office building, 12 Yuyan Road, Huangpu District, Guangzhou City, Guangdong Province Applicant after: PERCEPTION VISION MEDICAL TECHNOLOGY Co.,Ltd. Address before: 510275 No. 135 West Xingang Road, Guangzhou, Guangdong, Haizhuqu District Applicant before: SUN YAT-SEN University |
|
GR01 | Patent grant | ||
GR01 | Patent grant |