CN110866935A - Method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution - Google Patents
Method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution Download PDFInfo
- Publication number
- CN110866935A CN110866935A CN201810984658.9A CN201810984658A CN110866935A CN 110866935 A CN110866935 A CN 110866935A CN 201810984658 A CN201810984658 A CN 201810984658A CN 110866935 A CN110866935 A CN 110866935A
- Authority
- CN
- China
- Prior art keywords
- image
- radiotherapy
- false positive
- automatic segmentation
- radiotherapy structure
- 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
- 238000001959 radiotherapy Methods 0.000 title claims abstract description 61
- 230000011218 segmentation Effects 0.000 title claims abstract description 31
- 238000000034 method Methods 0.000 title claims abstract description 29
- 239000011159 matrix material Substances 0.000 claims abstract description 13
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 238000003062 neural network model Methods 0.000 claims abstract description 6
- 210000000056 organ Anatomy 0.000 claims abstract description 6
- 238000002679 ablation Methods 0.000 claims 1
- 238000013528 artificial neural network Methods 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 abstract description 3
- 210000004185 liver Anatomy 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 4
- 210000000920 organ at risk Anatomy 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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]
Abstract
The invention discloses a method, equipment and a storage medium for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution. The method comprises the following steps: acquiring a CT image of a patient and taking the image as a first image; predicting a human body radiotherapy structure mask by utilizing an automatic segmentation neural network model of a radiotherapy structure based on the first image to obtain an image as a second image; performing dot product on the matrix represented by the first image and the matrix of the second image, and intercepting the image of the area where the mask is located in the first image to obtain a third image; removing a false positive region of a radiotherapy structure in the third image according to the Hu value of the organ to obtain a fourth image; and performing edge extraction on the first image according to the fourth image to obtain an automatic delineation result for removing the false positive radiotherapy structure. The method can solve the problem that false positive exists in the automatic CT image delineation result based on the neural network. Thereby improving the precision of automatic segmentation and being beneficial to the accuracy and reliability of subsequent radiotherapy dose calculation.
Description
Technical Field
The invention belongs to the technical field of deep learning and radiotherapy, and relates to a method, equipment and a storage medium for removing false positives in radiotherapy structure automatic segmentation based on HU value distribution.
Background
In the process of carrying out radiotherapy to patient in the hospital, often involve the delineation of target area, at present the doctor mainly adopts the mode of manual delineation, and manual delineation is wasted time and energy, influences doctor's work efficiency, more influences patient's timely treatment. With the development of the deep learning technology, the automatic segmentation technology of medical images based on the deep learning technology becomes a popular research in the medical field. Automatic delineation of multiple human multi-part organs-at-risk and target areas has been developed and used in a number of domestic hospitals by Beijing Heart-connecting medical technology Limited. However, the delineation result of the existing automatic segmentation (the radiotherapy structure includes the target region and the organs at risk) often has a certain proportion of false positive, and the false positive is a place which is not the radiotherapy structure and is wrongly delineated into the radiotherapy structure when the radiotherapy structure in the CT image is delineated, so that the predicted radiotherapy structure during automatic delineation is larger than the actual radiotherapy structure, which has a great influence on the accuracy of the automatic segmentation result. Inaccurate automatic delineation will result in deviation in subsequent radiotherapy dose calculation, affecting the final radiotherapy effect.
Disclosure of Invention
The present invention is directed to a method, apparatus and storage medium for removing false positives in an automatic segmentation of radiotherapy structures based on HU value distribution, which overcome the disadvantages of the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme:
a method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution comprises the following steps:
(1) acquiring a CT image of a patient and taking the image as a first image;
(2) predicting a human body radiotherapy structure mask (mask) by utilizing an automatic segmentation neural network model of a radiotherapy structure based on the first image, and taking the obtained image as a second image;
(3) performing dot product on the matrix represented by the first image and the matrix of the second image, and intercepting the image of the area where the mask is located from the first image to obtain a third image;
(4) removing the false positive area of the radiotherapy structure in the third image according to the Hu value of the organ to obtain a fourth image;
(5) and performing edge extraction on the first image according to the fourth image to obtain an automatic delineation result for removing the false positive radiotherapy structure.
Further preferably, in step (2), the predicted radiotherapy structure mask of the human body is a binary image, the white area is the predicted radiotherapy structure area, and the black part is the background.
The fourth image is the binary mask image after the false positive is removed.
The present invention also provides a computing device comprising:
one or more processors;
a memory; and
one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for a method for removing false positives in an automatic segmentation of a radiotherapy structure based on a distribution of HU values.
The present invention also provides a computer readable storage medium storing one or more programs, the one or more programs comprising instructions adapted to be loaded from a memory and to perform the above-described method for removing false positives in an automatic segmentation of a radiotherapy structure based on a distribution of HU values.
The invention has the following beneficial results:
the method for removing the false positive in the radiotherapy structure automatic segmentation based on the HU value distribution can solve the problem that the false positive exists in the existing CT image automatic delineation result based on the neural network. Thereby improving the precision of automatic segmentation and being beneficial to the accuracy and reliability of subsequent radiotherapy dose calculation.
Drawings
FIG. 1 is a CT view of a liver to be delineated in an exemplary embodiment of the present invention.
Fig. 2 is a diagram of a liver mask image predicted by an automatic delineation model using an existing neural network in an exemplary embodiment of the invention, wherein white is the predicted liver region.
Fig. 3 is a graph showing the results of performing a matrix dot product on fig. 1 using fig. 2.
FIG. 4 is a diagram showing the improved mask result of the false positive processing performed on FIG. 3 by using the liver Hu value.
Fig. 5 is a sketch result (automatic sketch result after removing false positive) after extracting an edge using the improved mask obtained in fig. 4.
Fig. 6 is an automatic comparison of livers with and without false positive removed.
Detailed Description
In the prior art, a target area of a CT image can be automatically sketched by utilizing a Convolutional Neural Network (CNN), however, a false positive problem often occurs in an automatically sketched radiotherapy structure, and the area of the automatically sketched target area is larger than that of an actual radiotherapy structure. The invention provides a method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution to overcome the defects. The invention is further illustrated below with reference to the figures and examples.
Under normal conditions, the HU value of the CT image is distributed in the range of [ -1000,1000+ ], each organ at risk also has the own HU value distribution range [ x, y ], and the range is used for carrying out false positive elimination treatment on the prediction result. For example, in the case of an organ-at-risk liver, the HU values of the liver may be determined to be all greater than zero, and if the prediction result includes a negative value in the original image portion, it may be determined that a false positive portion exists in the prediction result.
A method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution comprises the following steps:
(1) acquiring a CT image of a patient and taking the image as a first image;
(2) predicting a mask image of a radiotherapy structure of a human body by utilizing an automatic segmentation neural network model of the radiotherapy structure based on the first image, and taking the mask image as a second image; the second image is a binary image, the white area is a predicted radiotherapy structure area, and the black part is a background;
(3) performing dot product on the matrix represented by the first image and the matrix represented by the second image, and intercepting the image of the area where the mask is located in the first image to obtain a third image;
(4) removing the false positive area of the radiotherapy structure in the third image according to the Hu value range of the organ to obtain a fourth image; the fourth image is a binary mask image with false positive removed;
(5) and performing edge extraction on the first image according to the fourth image to obtain an automatic delineation result of the radiotherapy structure without the false positive.
In an exemplary embodiment of the present invention, fig. 1 is a CT image (first image) of a human body to be delineated, the image mainly relating to a liver region of the human body; fig. 2 is a liver mask image (second image) predicted by using a conventional neural network automatic segmentation model, in which white is a predicted liver region, the image is a binarized image, the white region represents the predicted liver region, and has a value of 1, the black portion is a background, and the value is 0. Performing dot product on the matrixes respectively represented by the first image and the second image to obtain a third image, as shown in fig. 3, the third image is equivalent to extracting the liver part in fig. 1 based on the outline of the second image, and the HU value of the black part can be detected to be negative from the extracted image, so that a part of the corresponding second image is found to be a false positive region. And assigning the part with the HU value being negative in the third image as 0, and reassigning the part with the HU value being positive in the third image as 1, so as to obtain a fourth image shown in the figure 4, wherein the fourth image is a binary mask image with false positive removed. After edge extraction is performed on the first image by using the improved mask obtained in fig. 4, an automatic delineation result is obtained after the false positive is removed, as shown in fig. 5. Fig. 6 shows a comparison between the prediction result without the false positive processing and the prediction result after the false positive processing is completed, and it is found that a finer segmentation effect can be obtained by the false positive processing method provided by the present invention.
The present invention also provides a computing device comprising:
one or more processors;
a memory; and
one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for a method for removing false positives in an automatic segmentation of radiotherapy structures based on a distribution of HU values, the method comprising the steps of:
(1) acquiring a CT image of a patient and taking the image as a first image;
(2) predicting a human body radiotherapy structure mask by utilizing an automatic segmentation neural network model of a radiotherapy structure based on the first image to obtain an image as a second image;
(3) performing dot product on the matrix represented by the first image and the matrix of the second image, and intercepting the image of the area where the mask is located in the first image to obtain a third image;
(4) removing a false positive region of a radiotherapy structure in the third image according to the Hu value of the organ to obtain a fourth image;
(5) and performing edge extraction on the first image according to the fourth image to obtain an automatic delineation result for removing the false positive radiotherapy structure.
The present invention also provides a computer readable storage medium storing one or more programs, the one or more programs comprising instructions adapted to be loaded from a memory and to perform the method for removing false positives in an automatic segmentation of radiotherapy structures based on a distribution of HU values, the method comprising the steps of:
(1) acquiring a CT image of a patient and taking the image as a first image;
(2) predicting a human body radiotherapy structure mask by utilizing an automatic segmentation neural network model of a radiotherapy structure based on the first image to obtain an image as a second image;
(3) performing dot product on the matrix represented by the first image and the matrix of the second image, and intercepting the image of the area where the mask is located in the first image to obtain a third image;
(4) removing a false positive region of a radiotherapy structure in the third image according to the Hu value of the organ to obtain a fourth image;
(5) and performing edge extraction on the first image according to the fourth image to obtain an automatic delineation result for removing the false positive radiotherapy structure.
It should be understood that the various techniques described herein may be implemented in connection with hardware or software or, alternatively, with a combination of both. Thus, the methods and apparatus of the present invention, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine-readable storage medium, wherein, when the program is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the invention.
By way of example, and not limitation, computer readable media may comprise computer storage media and communication media. Computer storage media store information such as computer readable instructions, data structures, program modules or other data. Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. Combinations of any of the above are also included within the scope of computer readable media.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
As used herein, unless otherwise specified the use of the ordinal adjectives "first", "second", "third", etc., to describe a common object, merely indicate that different instances of like objects are being referred to, and are not intended to imply that the objects so described must be in a given sequence, either temporally, spatially, in ranking, or in any other manner.
Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.
The embodiments described above are intended to facilitate one of ordinary skill in the art in understanding and using the present invention. It will be readily apparent to those skilled in the art that various modifications to these embodiments may be made, and the generic principles described herein may be applied to other embodiments without the use of the inventive faculty. Therefore, the present invention is not limited to the embodiments described herein, and those skilled in the art should make improvements and modifications within the scope of the present invention based on the disclosure of the present invention.
Claims (5)
1. A method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution, adapted to be executed in a computing device, characterized in that: the method comprises the following steps:
(1) acquiring a CT image of a patient and taking the image as a first image;
(2) predicting a human body radiotherapy structure mask by utilizing an automatic segmentation neural network model of a radiotherapy structure based on the first image, and taking the obtained image as a second image;
(3) performing dot product on the matrix represented by the first image and the matrix represented by the second image, and intercepting the image of the area where the mask is located from the first image to obtain a third image;
(4) removing the false positive area of the radiotherapy structure in the third image according to the Hu value of the organ to obtain a fourth image;
(5) and performing edge extraction on the first image according to the fourth image to obtain an automatic delineation result of the radiotherapy structure without the false positive.
2. The method for removing false positives in radiotherapy structural automatic segmentation based on HU value distribution according to claim 1, wherein: in the step (2), the predicted human body radiotherapy structure mask is a binary image, a white area is a predicted radiotherapy structure area, and a black part is a background.
3. The method for removing false positives in radiotherapy structural automatic segmentation based on HU value distribution according to claim 1, wherein: the fourth image is the binary mask image after the false positive is removed.
4. A computing device, comprising:
one or more processors;
a memory; and
one or more programs stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for the method for automated segmentation of radiotherapy structures based on distribution of HU values according to any one of claims 1 to 3 above.
5. A computer readable storage medium storing one or more programs, the one or more programs comprising instructions adapted to be loaded from a memory and to perform the method for HU distribution based ablation of false positives in radiotherapy structured automatic segmentation as claimed in any one of the preceding claims 1 to 3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810984658.9A CN110866935A (en) | 2018-08-28 | 2018-08-28 | Method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810984658.9A CN110866935A (en) | 2018-08-28 | 2018-08-28 | Method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110866935A true CN110866935A (en) | 2020-03-06 |
Family
ID=69651332
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810984658.9A Pending CN110866935A (en) | 2018-08-28 | 2018-08-28 | Method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110866935A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117058172A (en) * | 2023-08-24 | 2023-11-14 | 吉林大学 | CT image multi-region segmentation method and device, electronic equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7149331B1 (en) * | 2002-09-03 | 2006-12-12 | Cedara Software Corp. | Methods and software for improving thresholding of coronary calcium scoring |
WO2016059385A1 (en) * | 2014-10-17 | 2016-04-21 | University College Cardiff Consultants Limited | A method for optimising segmentation of a tumour on an image, and apparatus therefor |
CN108447551A (en) * | 2018-02-09 | 2018-08-24 | 北京连心医疗科技有限公司 | A kind of automatic delineation method in target area based on deep learning, equipment and storage medium |
-
2018
- 2018-08-28 CN CN201810984658.9A patent/CN110866935A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7149331B1 (en) * | 2002-09-03 | 2006-12-12 | Cedara Software Corp. | Methods and software for improving thresholding of coronary calcium scoring |
WO2016059385A1 (en) * | 2014-10-17 | 2016-04-21 | University College Cardiff Consultants Limited | A method for optimising segmentation of a tumour on an image, and apparatus therefor |
CN108447551A (en) * | 2018-02-09 | 2018-08-24 | 北京连心医疗科技有限公司 | A kind of automatic delineation method in target area based on deep learning, equipment and storage medium |
Non-Patent Citations (2)
Title |
---|
ELMAR RENDON-GONZALEZ AND VOLODYMYR PONOMARYOV: ""Automatic Lung Nodule Segmentation and Classification in CT Images Based on SVM"", 《IEEE》 * |
张俊杰 等: ""基于空间分布的三维自动化肺结节分割算法"", 《数字视频》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117058172A (en) * | 2023-08-24 | 2023-11-14 | 吉林大学 | CT image multi-region segmentation method and device, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110310287B (en) | Automatic organ-at-risk delineation method, equipment and storage medium based on neural network | |
CN111291825B (en) | Focus classification model training method, apparatus, computer device and storage medium | |
CN107730507A (en) | A kind of lesion region automatic division method based on deep learning | |
US10628659B2 (en) | Intelligent tumor tracking system | |
CN110689521B (en) | Automatic identification method and system for human body part to which medical image belongs | |
CN107993220B (en) | Method and device for extracting vascular structure in X-ray radiography image | |
Men et al. | Automated quality assurance of OAR contouring for lung cancer based on segmentation with deep active learning | |
CN110287956A (en) | Vessel centerline automatic matching method and device | |
CN110827961A (en) | Automatic delineation method, device and storage medium for adaptive radiotherapy structure | |
CN106909947A (en) | CT image metal artifacts removing method and elimination system based on Mean Shift algorithms | |
CN112150472A (en) | Three-dimensional jaw bone image segmentation method and device based on CBCT (cone beam computed tomography) and terminal equipment | |
CN111724389B (en) | Method, device, storage medium and computer equipment for segmenting CT image of hip joint | |
CN111145160A (en) | Method, device, server and medium for determining coronary artery branch where calcified area is located | |
CN111353978B (en) | Method and device for identifying heart anatomy structure | |
Song et al. | Adaptive fast marching method for automatic liver segmentation from CT images | |
CN111568451A (en) | Exposure dose adjusting method and system | |
CN110866935A (en) | Method for removing false positive in radiotherapy structure automatic segmentation based on HU value distribution | |
CN113689938B (en) | Medical image sketching method, device, storage medium and processor | |
CN107767961B (en) | Bone template design method for bone fracture plate type discrimination | |
CN114494189A (en) | Nursing wound image recognition and detection method and device | |
CN113344926A (en) | Method, device, server and storage medium for recognizing biliary-pancreatic ultrasonic image | |
CN111738975B (en) | Image identification method and image identification device | |
CN116309647B (en) | Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device | |
CN108877927A (en) | A kind of medical image diagnosis method | |
CN116883372A (en) | Method and system for adaptively identifying tumor based on blood vessel region image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200306 |