CN115100155A - Method and system for establishing radiation pneumonitis prediction model - Google Patents
Method and system for establishing radiation pneumonitis prediction model Download PDFInfo
- Publication number
- CN115100155A CN115100155A CN202210765929.8A CN202210765929A CN115100155A CN 115100155 A CN115100155 A CN 115100155A CN 202210765929 A CN202210765929 A CN 202210765929A CN 115100155 A CN115100155 A CN 115100155A
- Authority
- CN
- China
- Prior art keywords
- lung
- ctv
- dose
- omics
- lung tissue
- 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.)
- Granted
Links
- 206010037765 Radiation pneumonitis Diseases 0.000 title claims abstract description 74
- 238000000034 method Methods 0.000 title claims abstract description 43
- 210000004072 lung Anatomy 0.000 claims abstract description 126
- 230000005855 radiation Effects 0.000 claims abstract description 45
- 238000001959 radiotherapy Methods 0.000 claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 238000007637 random forest analysis Methods 0.000 claims abstract description 22
- 206010058467 Lung neoplasm malignant Diseases 0.000 claims abstract description 21
- 201000005202 lung cancer Diseases 0.000 claims abstract description 21
- 208000020816 lung neoplasm Diseases 0.000 claims abstract description 21
- 238000012216 screening Methods 0.000 claims abstract description 18
- 238000003384 imaging method Methods 0.000 claims abstract description 13
- 238000001914 filtration Methods 0.000 claims description 25
- 238000012512 characterization method Methods 0.000 claims description 14
- 239000013598 vector Substances 0.000 claims description 14
- 206010028980 Neoplasm Diseases 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 11
- 238000003745 diagnosis Methods 0.000 abstract description 8
- 238000012545 processing Methods 0.000 description 10
- 238000003066 decision tree Methods 0.000 description 8
- 230000011218 segmentation Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000003902 lesion Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 208000000059 Dyspnea Diseases 0.000 description 5
- 206010013975 Dyspnoeas Diseases 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 5
- 201000010099 disease Diseases 0.000 description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 5
- 208000004852 Lung Injury Diseases 0.000 description 4
- 206010069363 Traumatic lung injury Diseases 0.000 description 4
- 230000001154 acute effect Effects 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000004907 flux Effects 0.000 description 4
- 231100000515 lung injury Toxicity 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 206010011224 Cough Diseases 0.000 description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000013145 classification model Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000011404 fractionated radiotherapy Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 229910052760 oxygen Inorganic materials 0.000 description 3
- 239000001301 oxygen Substances 0.000 description 3
- 208000024891 symptom Diseases 0.000 description 3
- 206010006187 Breast cancer Diseases 0.000 description 2
- 208000026310 Breast neoplasm Diseases 0.000 description 2
- 230000000954 anitussive effect Effects 0.000 description 2
- 229940124584 antitussives Drugs 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000009795 derivation Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 206010008479 Chest Pain Diseases 0.000 description 1
- 206010025323 Lymphomas Diseases 0.000 description 1
- 206010035664 Pneumonia Diseases 0.000 description 1
- 206010037660 Pyrexia Diseases 0.000 description 1
- 208000004756 Respiratory Insufficiency Diseases 0.000 description 1
- 230000008382 alveolar damage Effects 0.000 description 1
- 230000003444 anaesthetic effect Effects 0.000 description 1
- 239000003242 anti bacterial agent Substances 0.000 description 1
- 229940088710 antibiotic agent Drugs 0.000 description 1
- 239000003434 antitussive agent Substances 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007596 consolidation process Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- ZZVUWRFHKOJYTH-UHFFFAOYSA-N diphenhydramine Chemical compound C=1C=CC=CC=1C(OCCN(C)C)C1=CC=CC=C1 ZZVUWRFHKOJYTH-UHFFFAOYSA-N 0.000 description 1
- 208000017574 dry cough Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000003176 fibrotic effect Effects 0.000 description 1
- 230000036732 histological change Effects 0.000 description 1
- 229940088597 hormone Drugs 0.000 description 1
- 239000005556 hormone Substances 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 231100000516 lung damage Toxicity 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000003533 narcotic effect Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000002685 pulmonary effect Effects 0.000 description 1
- 208000005069 pulmonary fibrosis Diseases 0.000 description 1
- 230000000693 radiobiological effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 201000004193 respiratory failure Diseases 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 150000003431 steroids Chemical class 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
- 210000000115 thoracic cavity Anatomy 0.000 description 1
- 238000009423 ventilation Methods 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
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
- A61N5/103—Treatment planning systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
- A61N5/103—Treatment planning systems
- A61N5/1031—Treatment planning systems using a specific method of dose optimization
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; 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
- 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
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
- A61N5/00—Radiation therapy
- A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
- A61N5/1048—Monitoring, verifying, controlling systems and methods
- A61N5/1049—Monitoring, verifying, controlling systems and methods for verifying the position of the patient with respect to the radiation beam
- A61N2005/1061—Monitoring, verifying, controlling systems and methods for verifying the position of the patient with respect to the radiation beam using an x-ray imaging system having a separate imaging source
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30061—Lung
-
- 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
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Public Health (AREA)
- Databases & Information Systems (AREA)
- Veterinary Medicine (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Pathology (AREA)
- Molecular Biology (AREA)
- Animal Behavior & Ethology (AREA)
- Multimedia (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The application belongs to the technical field of medical intelligence, and relates to a method for establishing a radiation pneumonitis prediction model, which comprises the following steps: identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV to obtain LUNG tissue LUNG-CTV without the clinical target area CTV; acquiring three-dimensional physical dose of LUNG tissue LUNG-CTV region in a radiotherapy plan of a LUNG cancer patient, and converting the three-dimensional physical dose into 2Gy fractionated radiation equivalent biological dose; extracting the imaging omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of 2Gy fractionated radiation equivalent biological dose; and (4) carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm. The application also provides a system and equipment for establishing the radiation pneumonitis prediction model and a storage medium thereof. In addition, the application also relates to a block chain technology, and the result of the radiation pneumonitis diagnosis can be stored in the block chain.
Description
Technical Field
The application relates to the technical field of medical intelligence, in particular to a method, a system and equipment for establishing a radiation pneumonitis prediction model and a storage medium thereof.
Background
Radiation Therapy (RT) is one of the commonly used methods for treating tumors, and over 70% of lung cancer patients need Radiation therapy during the treatment process. Since sometimes a large volume of lung tissue needs to be exposed to high dose Radiation during treatment, the lung tissue surrounding the tumor is exposed to Radiation doses exceeding its tolerance threshold, thereby causing different degrees of Radiation lung damage (Radiation lung Injury), which can affect the quality of life and therapeutic effect of the patient, and even lead to death in the serious cases. Therefore, radiation-induced lung injury is the most major limiting factor and most common complication of breast tumor radiotherapy. According to statistics of data at home and abroad, about 5-15% of patients undergoing radiotherapy for lymphoma, lung cancer and breast cancer have radiation pneumonitis with different degrees, the clinical manifestations are low fever and nonspecific respiratory symptoms, severe patients have symptoms such as dyspnea and chest pain, and the imaging shows that the interstitial density of lung is increased; after the acute phase, approximately 80% of patients will continue to develop histological changes, and fibrotic lesions will appear; patients with symptomatic radiation pneumonitis are generally treated clinically with oxygen, hormones, antibiotics and the like, and if the patients do not respond to the treatment, the patients may die of the radiation pneumonitis. Radiation lung injury is therefore the major dose limiting factor in thoracic Radiation therapy, and high doses of Radiation given to healthy lung tissue can cause alveolar damage, leading to acute Radiation Pneumonitis (RP). Once the radiation pneumonitis occurs, it is difficult to reverse, so the prevention is more important than the treatment.
In recent studies of radiation pneumonitis, DVH plots yield measurements such as Mean Lung Dose (MLD) or percent lung volume (V) above a prescribed threshold dose dose ) While this is related to the development of radiation pneumonitis, when the incidence of radiation pneumonitis is predicted on the basis of a certain study, the accuracy is insufficient and the prediction is difficult.
Disclosure of Invention
The embodiment of the application aims to provide a method, a system, equipment and a storage medium for establishing a radiation pneumonitis prediction model so as to solve the technical problem that incidence of radiation pneumonitis is difficult to predict in the prior art.
In order to solve the above technical problem, an embodiment of the present application provides a method for establishing a radiation pneumonitis prediction model, which adopts the following technical scheme: the method comprises the following steps:
identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed;
acquiring three-dimensional physical dose of LUNG tissue LUNG-CTV region in a radiotherapy plan of a LUNG cancer patient, and converting the three-dimensional physical dose into 2Gy fractionated radiation equivalent biological dose through an EQD2 formula;
extracting the imaging omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of 2Gy fractionated radiation equivalent dose;
carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm, and filtering the features representing redundant information;
combining the filtered image omics characteristics and the filtered dose omics characteristics into multi-modal omics characteristics and inputting the multi-modal omics characteristics into different classifiers;
and extracting multi-modal omics characteristics with optimal performance in each class, and establishing a radiation pneumonitis prediction model according to the multi-modal omics characteristics with optimal performance.
In order to solve the above technical problem, an embodiment of the present application further provides a system for building a radiation pneumonitis prediction model, where the system includes:
the identification module is used for identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed;
a transformation module for obtaining three-dimensional physical dose of LUNG tissue Lung-CTV region in radiotherapy planning of LUNG cancer patient, and passing through EQD 2 Converting the three-dimensional physical dose into 2Gy fractionated radiation equivalent biological dose by a formula;
the grading module is used for extracting the imaging omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of 2Gy graded radiation equivalent dose;
the filtering module is used for performing feature screening on the image omics features and the dose omics features by using a random forest algorithm and filtering the features representing redundant information;
the classification module is used for forming multi-modal omics characteristics by the filtered image omics characteristics and the filtered dose omics characteristics and inputting the multi-modal omics characteristics into different classifiers;
and the establishing module is used for extracting the multimodality omics characteristics with optimal performance in each class and establishing the radiation pneumonitis prediction model according to the multimodality characteristics with optimal performance.
In order to solve the above technical problem, an embodiment of the present application further provides a computer device, which adopts the following technical solutions: comprising a memory having computer readable instructions stored therein and a processor that when executed implements the steps of the method of creating a predictive model of radiation pneumonitis as described above.
In order to solve the above technical problem, an embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solutions: the computer readable storage medium has stored thereon computer readable instructions which, when executed by a processor, implement the steps of the method of building a prediction model of radiation pneumonitis as described above.
Compared with the prior art, the embodiment of the application mainly has the following beneficial effects: identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed; acquiring three-dimensional physical dose of LUNG tissue LUNG-CTV region in a radiotherapy plan of a LUNG cancer patient, converting the three-dimensional physical dose into 2Gy fractional radiation equivalent biological dose through an EQD2 formula, and extracting image omics characteristics of the LUNG tissue LUNG-CTV region and 2Gy fractional radiation equivalent dose omics characteristics; carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm, and filtering the features representing redundant information; combining the filtered image omics characteristics and the filtered dose omics characteristics into multi-modal omics characteristics and inputting the multi-modal omics characteristics into different classifiers; and extracting multi-modal omics characteristics with optimal performance in each class, and establishing a radiation pneumonitis prediction model according to the multi-modal omics characteristics with optimal performance. Accurate identification and segmentation of clinical target areas CTV are realized, automatic diagnosis, positioning and filtering analysis of lesion features of LUNG tissues LUNG are facilitated, intelligent identification of radiation pneumonitis and disease prediction are realized, judgment efficiency of radiation pneumonitis is improved, and medical safety of critical patients is guaranteed.
Drawings
In order to more clearly illustrate the solution of the present application, the drawings used in the description of the embodiments of the present application will be briefly described below, and it is obvious that the drawings in the description below are some embodiments of the present application, and that other drawings may be obtained by those skilled in the art without inventive effort.
FIG. 1 is an exemplary system architecture diagram to which the present application may be applied;
FIG. 2 is a flow diagram of one embodiment of a method of building a prediction model of radiation pneumonitis;
FIG. 3 is a diagram of a CT image processing variation process;
FIG. 4 is a representation of an example of feature extraction;
FIG. 5 is a schematic diagram of an embodiment of a system for building a prediction model of radiation pneumonitis;
FIG. 6 is a schematic block diagram of one embodiment of a computer device according to the present application.
Detailed Description
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "including" and "having," and any variations thereof, in the description and claims of this application and the description of the above figures are intended to cover non-exclusive inclusions. The terms "first," "second," and the like in the description and claims of this application or in the above-described drawings are used for distinguishing between different objects and not for describing a particular order.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein may be combined with other embodiments.
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.
As shown in fig. 1, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. Various communication client applications, such as a web browser application, a knowledge application, an instant messaging tool, a mailbox client, social platform software, and the like, may be installed on the terminal devices 101, 102, 103.
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, mpeg compression standard Audio Layer 3), MP4 players (Moving Picture Experts Group Audio Layer IV, mpeg compression standard Audio Layer 4), laptop portable computers, desktop computers, and the like.
The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the terminal devices 101, 102, 103.
It should be noted that the method for building the radiation pneumonitis prediction model provided in the embodiment of the present application is generally executed by a server, and accordingly, the system for building the radiation pneumonitis prediction model is generally disposed in the server.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
With continuing reference to figures 2-4, figure 2 illustrates a flow diagram of one embodiment of a method of building a radiation pneumonitis prediction model according to the present application. The method for establishing the radiation pneumonitis prediction model comprises the following steps:
step S201, identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed;
on one hand, the lung CT image is detected and obtained by a patient in real time, and on the other hand, the lung CT image can be extracted from a pre-stored radiation pneumonitis database, wherein the radiation pneumonitis database is established by acquiring the CT image of the lung of a clinical hospital, asking a doctor to mark a radiation pneumonitis area and establishing the radiation pneumonitis database through computer training and learning.
Calculating an edge optimal threshold of the LUNG tissue Lung and an edge optimal threshold of the clinical target area CTV by recognizing the LUNG tissue LUNG and the clinical target area CTV in the CT image; determining the contour edge of the LUNG tissue LUNG and the contour edge of the clinical target area CTV in the CT image based on the edge optimal threshold of the LUNG tissue and the edge optimal threshold of the clinical target area CTV; segmenting LUNG tissue LUNG and clinical target region CTV, and obtaining LUNG tissue LUNG-CTV after removing clinical target region CTV. The edge optimal threshold refers to an edge covered by a lesion occurrence region in a clinical target region CTV, specifically, a threshold of a gray value is calculated by calculating an average value and a standard deviation of all pixels of a CT image and then by a formula: t is μ + z σ,
wherein T is the threshold of the gray value of the pixel, mu is the average gray value of all the pixels, sigma is the standard deviation of all the pixels, D is the number of the pixels of the standard CT image, x i For the gray value of each pixel, z is a coefficient constant, and the value range is as follows: -5.0, generally 1.7 or 1.8, and the specific value is actually determined according to the situation.
Further, the step of examining the LUNG tissue Lung-CTV obtained after the segmentation to see if it further contains the clinical target volume CTV comprises: identifying a clinical target region CTV in LUNG tissue LUNG-CTV; when no clinical target area CTV exists in the identified LUNG tissue LUNG-CTV, optimizing the edge of the LUNG tissue LUNG-CTV through mathematical morphology; when the LUNG tissue Lung-CTV is identified to have the clinical target CTV, the step of calculating the marginal optimal threshold of the LUNG tissue Lung and the marginal optimal threshold of the clinical target CTV is performed. And (3) based on mathematical morphology, measuring the elements of the morphological structure of the clinical target region CTV and extracting the corresponding shape in the CT image, thereby identifying the clinical target region CTV in the LUNG tissue Lung-CTV. Naturally, the CT image needs to be preprocessed before being processed by using mathematical morphology, and other interference factors of the CT image are eliminated by a preprocessing mode, specifically including image gray processing, image median filtering processing, and image histogram equalization processing.
Preferably, in this embodiment, the CT image is mainly subjected to image grayscale processing, the grayscale processed image only includes luminance information but not color information, the grayscale image is convenient to store, and the processing efficiency can also be improved.
Step S202, acquiring radiotherapy flux information of a lung cancer patient, and converting flux in the radiotherapy flux information into 2Gy fractionated radiation equivalent dose through an EQD2 formula;
specifically, acquiring a radiation therapy dose file of a lung cancer patient from a planning system; extracting three-dimensional physical dose of LUNG tissue LUNG-CTV region in a LUNG cancer patient radiotherapy plan based on Python programming; using EQD 2 The formula converts the three-dimensional physical dose into a 2Gy fractionated radiation equivalent biological dose.
Equivalent Dose of 2Gy fractionated radiation, i.e., Equivalent biological Dose (Equivalent Dose in 2Gy/F, EQD) Equivalent to conventional 2Gy fractionated radiation 2 ). With EQD 2 And (3) as a standard equivalent dose, calculating the equivalent dose of the fractionated radiotherapy, such as the super-fractionated and large-fractionated radiotherapy technology, or changing the conventional radiotherapy into the super-fractionated radiotherapy for 2 times a day after a certain dose. Sometimes, the EQD of the patient can be calculated when the radiotherapy is interrupted for several days to several weeks due to the patient or equipment repair and the like 2 Equivalent dosage values, and thus direct further treatment. The LQ Model (Linear predictive Model, LQ) is derived from direct derivation of cell survival curves, and equivalent doses and mathematical derivation formulas based on LQ models facilitate standardized comparison of biological doses at different time-dose-times or different segmentation sequences. EQD 2 The calculation formula is as follows:
wherein D represents the total dose and D is the divided dose. The α/β value reflects the radiobiological properties of the tissue, β 0/β 1 being larger (e.g., when α/β is 10Gy or more) for an equivalent biological dose equivalent to early response tissue upon irradiation by a conventional radiotherapy protocol (D/N is 2Gy), while α/β being smaller (e.g., when α/β is 3Gy or less) for an equivalent biological dose equivalent to late response tissue upon irradiation by a conventional radiotherapy protocol (D/N is 2Gy), typically the α/β of LUNG tissue LUNG is between 4 and 5. In this embodiment, the three-dimensional physical dose of LUNG tissue LUNG-CTV region in a patient radiation treatment plan is extracted based on Python programming, and then EQD is utilized 2 The formula converts the three-dimensional physical dose to an equivalent biological dose of conventional 2Gy fractionated radiation.
Step S203, extracting the imaging omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of 2Gy fractionated radiation equivalent dose;
specifically, an image omics feature of LUNG tissue Lung-CTV is extracted based on Python programming, wherein the image omics feature comprises at least one feature of tumor shape, gray level distribution, texture characteristic, wavelet and Fourier, wherein the tumor shape, gray level distribution and texture characteristic are low-order features, and are transformed into high-order features in the modes of wavelet or Fourier and the like;
respectively extracting the iconomics characteristics of the Lung tissue LUNG-CTV region in the CT image and the image after wavelet change based on Python programming, and extracting the EQD of the Lung tissue LUNG-CTV region in the CT image and the image after wavelet change 2 And the dose omics characteristics under the mode comprise 863 characteristics, such as first-order statistics, 2D and 3D-based shape characteristics, gray level co-occurrence matrixes, gray level running length matrixes, gray level region size matrixes, gray level dependency matrixes, neighborhood gray level difference matrixes and the like.
Step S204, performing feature screening on the image omics features and the dose omics features by using a random forest algorithm, and filtering the features representing redundant information;
the random forest algorithm generates a model by training a plurality of decision trees and then votes by comprehensively utilizing the classification results of the decision trees, thereby realizing classification. The random forest algorithm only needs the number t of decision trees constructed by two parameters, and the number of input features to be considered when each node of the decision trees is split. The more important hyper-parameters in the random forest model comprise the number of decision trees in the random forest, the maximum depth of the decision trees and the minimum sample number required for dividing internal nodes, so that the three parameters are used as hyper-parameters for optimizing the model when the random forest model is established. In addition, the random forest algorithm can return characteristic variables, the average value is taken from the regression results and is used as the final calculation result, and important variables in the characteristic set can be screened based on the characteristic importance.
Specifically, a random forest algorithm is used for carrying out feature screening on the image omics features and the dose omics features, obtaining the features representing redundant information and the features representing vector information, filtering the features representing redundant information, sequentially arranging the features representing vector information, obtaining the features representing vector information higher than a preset representation vector, and setting the features as multi-modal omics features. The characterization learning is performed by learning the use characteristics through a computer and learning how to extract the characteristics at the same time, the characterization learning is performed through a neural network and a multi-layer perceptron supervision computer, the characteristics of characterization redundant information are obtained after the characterization learning, and the characteristics of a plurality of characterization vector information are sequentially arranged, so that the characterization redundant information can be conveniently removed, the most valuable characteristics can be conveniently screened, the information characterization capability is enhanced, and the characterization vectors are preset in the embodiment according to the actual situation.
Step S205, combining the filtered image omics characteristics and the filtered dose omics characteristics into multi-modal omics characteristics and inputting the multi-modal omics characteristics into different classifiers;
the classifier classifies the multimodality omics characteristics in data mining and applies algorithms such as decision trees, logistic regression, naive Bayes, neural networks and the like. Specifically, multimodality omics features are used as samples (including positive samples and negative samples), all samples are divided into two parts, namely training samples and testing samples, a classifier algorithm is executed on the training samples to generate classification models, the classification models are executed on the testing samples to generate prediction results, evaluation indexes are calculated according to the prediction results, and the performance of the evaluation classification models is determined according to the evaluation indexes. In this embodiment, the screened image omics features and dose omics features are trained by using 11 classifiers (AdaBoost, multinomial nb, gaussian nb, pasiveagnesive, kneighbos, randomfort, DecisionTree, Perceptron, SVM, Ridge, logistic regression), and five-fold cross validation is adopted.
Preferably, before the classifier classifies, the multimodality omic features are classified firstly through a feature selection algorithm, specifically, data are selected from the multimodality omic features for the classifier to use, and N optimal features are selected from the existing original features as specific features to reduce the dimensionality of the data set, and of course, random search or heuristic search can also be performed.
Specifically, based on a feature selection algorithm, multi-modal omics features are classified; and (4) performing iterative learning on the multimodality omic features by using a classifier, and extracting the multimodality omic features with the optimal performance in each class.
And S206, extracting multimodality omics characteristics with optimal performance in each category, and establishing a radiation pneumonitis prediction model according to the multimodality characteristics with optimal performance.
Grading in the patient's radiolucent pneumonia by reference to the RTOG acute radiation lung injury grading criteria, as shown in detail: level 0, no change; grade 1, dyspnea with mild dry cough or exertion; 2, continuous cough, need of anesthetic antitussive, dyspnea with slight movement, but no dyspnea at rest; grade 3, severe cough, ineffective narcotic antitussives, or dyspnea at rest, clinical or imaging evidence of acute radiation pneumonitis, intermittent oxygen inhalation or possible need for steroid therapy; grade 4, severe respiratory insufficiency, continuous oxygen inhalation or assisted ventilation treatment; grade 5, directly dying from the radiation reaction. (specifically, grading is performed by imaging examination within 6 months after the clinical symptoms and radiotherapy of the patient are finished (showing that the irradiation field shows the occurrence of the piloedema or the consolidation, and except pulmonary infection, tumor recurrence and radiation pulmonary fibrosis, which can be included in the radiation pneumonitis)) the observation end point of the embodiment is equal to or more than 2-grade radiation pneumonitis, the label of the radiation pneumonitis equal to or less than 1 grade is 0, and the label of the radiation pneumonitis equal to or more than 2 grade is 1.
And evaluating the effectiveness of the model by using the Accuracy (ACC), the False Positive Rate (FPR) and the AUC value of the area under the ROC curve as indexes. Accuracy (Accuracy) is the ratio of how many judgments are correct among all judgments, i.e., a positive judgment is positive and a negative judgment is negative among all judgments. Precision (Precision) refers to how many of the features predicted to be positive are correct. The AUC value of the area under the ROC curve refers to the random probability of the score of a positive class feature being greater than the score of a negative class feature (score refers to the scoring of a classifier) for a positive class feature and a negative class feature. False Positive Rate (FPR) refers to the number of detected true positive features divided by the number of all true positive features.
The embodiment mainly has the following beneficial effects: identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting LUNG and CTV to obtain LUNG tissue LUNG-CTV after removing clinical target area CTV; obtaining three-dimensional physical dose of LUNG tissue Lung-CTV region in radiotherapy plan of LUNG cancer patient, and passing through EQD 2 Converting the three-dimensional physical dose into 2Gy fractional radiation equivalent biological dose by a formula, and extracting the imaging omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of the 2Gy fractional radiation equivalent dose; carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm, and filtering the features representing redundant information; combining the filtered image omics characteristics and the filtered dose omics characteristics into multi-modal omics characteristics and inputting the multi-modal omics characteristics into different classifiers; and extracting multi-modal omics characteristics with optimal performance in each class, and establishing a radiation pneumonitis prediction model according to the multi-modal omics characteristics with optimal performance. Accurate identification and segmentation of clinical target areas CTV are realized, automatic diagnosis, positioning and filtering analysis of lesion features of LUNG tissues LUNG are facilitated, intelligent identification of radiation pneumonitis and disease prediction are realized, judgment efficiency of radiation pneumonitis is improved, and medical safety of critical patients is guaranteed.
It is emphasized that, in order to further ensure the privacy and safety of the diagnosis result of the radiation pneumonitis, the diagnosis result of the radiation pneumonitis can also be stored in a node of a block chain.
The block chain referred by the application is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a string of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, which is used for verifying the validity (anti-counterfeiting) of the information and generating a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware associated with computer readable instructions, which can be stored in a computer readable storage medium, and when executed, the processes of the embodiments of the methods described above can be included. The storage medium may be a non-volatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Access Memory (RAM).
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and may be performed in other orders unless otherwise indicated herein. Moreover, at least a portion of the steps in the flow chart of the figure may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of execution is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
With further reference to fig. 5, as an implementation of the method shown in fig. 2-4, the present application provides an embodiment of a system for building a prediction model of radiation pneumonitis, which corresponds to the embodiment of the method shown in fig. 2-4.
As shown in fig. 5, the system 300 for building a prediction model of radiation pneumonitis according to the present embodiment includes: an identification module 301, a conversion module 302, a classification module 303, a filtering module 304, a classification module 305, and an establishment module 306. Wherein:
the identification module 301 is configured to identify LUNG tissue LUNG and clinical target area CTV in the CT image, and segment the LUNG and CTV to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed;
a transformation module 302 for obtaining a three-dimensional physical dose of LUNG tissue Lung-CTV region in a radiation therapy plan for a patient with LUNG cancer, and passing the dose through an EQD 2 Converting the three-dimensional physical dose into 2Gy fractionated radiation equivalent biological dose by a formula;
the grading module 303 is used for extracting imaging omics characteristics of LUNG tissue Lung-CTV regions and dose omics characteristics of 2Gy graded radiation equivalent dose;
the filtering module 304 is configured to perform feature screening on the image omics features and the dose omics features by using a random forest algorithm, and filter features representing redundant information;
the classification module 305 is configured to combine the filtered image omics features and the filtered dose omics features into multi-modal omics features, and input the multi-modal omics features into different classifiers;
the establishing module 306 is used for extracting multimodality omics characteristics with optimal performance in each category and establishing the radiation pneumonitis prediction model according to the multimodality characteristics with optimal performance.
The embodiment mainly has the following beneficial effects: identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed; obtaining three-dimensional physical dose of LUNG tissue Lung-CTV region in radiotherapy plan of LUNG cancer patient, and passing through EQD 2 Converting the three-dimensional physical dose into 2Gy fractional radiation equivalent dose by a formula, and extracting the image omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of the 2Gy fractional radiation equivalent dose; carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm, and filtering the features representing redundant information; combining the filtered image omics characteristics and the filtered dose omics characteristics into multi-modal omics characteristics and inputting the multi-modal omics characteristics into different classifiers; and extracting multi-modal omics characteristics with optimal performance in each class, and establishing a radiation pneumonitis prediction model according to the multi-modal omics characteristics with optimal performance. Accurate identification and segmentation of clinical target areas CTV are realized, automatic diagnosis, positioning and filtering analysis of lesion features of LUNG tissues LUNG are facilitated, intelligent identification of radiation pneumonitis and disease prediction are realized, judgment efficiency of radiation pneumonitis is improved, and medical safety of critical patients is guaranteed.
In some optional implementations of this embodiment, the identifying module 301 includes:
the identification unit is used for identifying LUNG tissues LUNG and clinical target areas CTV in the CT image;
a calculation unit for calculating a marginal optimum threshold for LUNG tissue Lung and a marginal optimum threshold for a clinical target volume CTV;
a determining unit, configured to determine a contour edge of the LUNG tissue LUNG and a contour edge of the clinical target area CTV in the CT image based on the edge optimal threshold of the LUNG tissue LUNG and the edge optimal threshold of the clinical target area CTV;
and the segmentation unit is used for segmenting the LUNG tissue LUNG and the clinical target area CTV and obtaining the LUNG tissue LUNG-CTV after the clinical target area CTV is removed.
In some optional implementations of this embodiment, the system 300 further includes:
and an examination module for identifying a clinical target area CTV in the LUNG tissue LUNG-CTV, wherein if no clinical target area CTV is identified in the LUNG tissue LUNG-CTV, the border of the LUNG tissue LUNG-CTV is optimized by mathematical morphology, and wherein if the clinical target area CTV is identified in the LUNG tissue LUNG-CTV, the step of calculating a border optimal threshold for the LUNG tissue LUNG and a border optimal threshold for the clinical target area CTV is performed.
In some optional implementations of this embodiment, the conversion module 302 includes:
an acquisition unit for acquiring a radiation therapy dose file of a lung cancer patient from a planning system;
a flux acquisition unit for extracting a three-dimensional physical dose of a LUNG tissue Lung-CTV region in a radiation therapy plan of a LUNG cancer patient based on Python programming;
a conversion unit for utilizing EQD 2 The formula converts the three-dimensional physical dose into a 2Gy fractionated radiation equivalent dose.
In some optional implementations of this embodiment, the classifying module 303 includes:
the image extraction unit is used for extracting the image omics characteristics of the LUNG tissue Lung-CTV based on Python programming, wherein the image omics characteristics comprise at least one of the characteristics of tumor shape, gray distribution, texture characteristics, wavelet passing and Fourier;
the dose extraction unit is used for extracting dose omics characteristics in 2Gy fractionated radiation equivalent dose based on Python programming, wherein the dose omics characteristics comprise at least one of first-order statistics, 2D and 3D-based shape characteristics, gray level co-occurrence matrixes, gray level operation length matrixes, gray level area size matrixes, gray level dependency matrixes, neighborhood gray level difference matrixes and the like.
In some optional implementations of this embodiment, the filtering module 304 includes:
the screening unit is used for carrying out feature screening on the image omics features and the dose omics features by utilizing a random forest algorithm;
the characteristic obtaining unit is used for obtaining the characteristics representing the redundant information and the characteristics representing the vector information;
the filtering unit is used for filtering the characteristics representing the redundant information and sequentially arranging the characteristics representing the vector information;
and the setting unit is used for acquiring the characteristics of the characterization vector information higher than the preset characterization vector and setting the characteristics as multimodality omics characteristics.
In some optional implementations of this embodiment, the classification module 305 includes:
the classification unit is used for classifying multimodality omics characteristics based on a characteristic selection algorithm;
and the iteration unit is used for performing iterative learning on the multimodality omic features by using the classifier and extracting the multimodality omic features with optimal performance in each class.
In order to solve the technical problem, an embodiment of the present application further provides a computer device. Referring to fig. 6, fig. 6 is a block diagram of a basic structure of a computer device according to the present embodiment.
The computer device 4 comprises a memory 41, a processor 42, a network interface 43 communicatively connected to each other via a system bus. It is noted that only computer device 4 having components 41-43 is shown in FIG. 6, but it is understood that not all of the illustrated components are required to be implemented, and that more or fewer components may alternatively be implemented. As will be understood by those skilled in the art, the computer device is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and the hardware includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like.
The computer device can be a desktop computer, a notebook, a palm computer, a cloud server and other computing devices. The computer equipment can carry out man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch panel or voice control equipment and the like.
The memory 41 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as a hard disk or a memory of the computer device 6. In other embodiments, the memory 41 may also be an external storage device of the computer device 6, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, provided on the computer device 4. Of course, the memory 41 may also include both internal and external storage devices of the computer device 4. In this embodiment, the memory 41 is generally used for storing an operating system installed in the computer device 4 and various types of application software, such as computer readable instructions of a method for building a radiation pneumonitis prediction model. Further, the memory 41 may also be used to temporarily store various types of data that have been output or are to be output.
The processor 42 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is configured to execute computer readable instructions stored in the memory 41 or process data, such as computer readable instructions for executing the method for building the radiation pneumonitis prediction model.
The network interface 43 may comprise a wireless network interface or a wired network interface, and the network interface 43 is generally used for establishing communication connection between the computer device 4 and other electronic devices.
The present embodiment has the following advantagesThe effect is as follows: identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed; obtaining three-dimensional physical dose of LUNG tissue Lung-CTV region in radiotherapy plan of LUNG cancer patient, and passing through EQD 2 Converting the three-dimensional physical dose into 2Gy fractional radiation equivalent dose by a formula, and extracting the image omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of the 2Gy fractional radiation equivalent dose; carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm, and filtering the features representing redundant information; combining the filtered image omics characteristics and the filtered dose omics characteristics into multi-modal omics characteristics and inputting the multi-modal omics characteristics into different classifiers; and extracting multi-modal omics characteristics with optimal performance in each class, and establishing a radiation pneumonitis prediction model according to the multi-modal omics characteristics with optimal performance. Accurate identification and segmentation of clinical target areas CTV are realized, automatic diagnosis, positioning and filtering analysis of lesion features of LUNG tissues LUNG are facilitated, intelligent identification of radiation pneumonitis and disease prediction are realized, judgment efficiency of radiation pneumonitis is improved, and medical safety of critical patients is guaranteed.
The present application further provides another embodiment, which is to provide a computer-readable storage medium storing computer-readable instructions executable by at least one processor to cause the at least one processor to perform the steps of the method for building a prediction model of radiation pneumonitis as described above.
The embodiment mainly has the following beneficial effects: identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed; obtaining three-dimensional physical dose of LUNG tissue Lung-CTV region in radiotherapy plan of LUNG cancer patient, and passing through EQD 2 Converting the three-dimensional physical dose into 2Gy fractional radiation equivalent biological dose by a formula, and extracting the imaging omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of the 2Gy fractional radiation equivalent dose; carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm, and carrying out feature screening on the features representing redundant informationFiltering; combining the filtered image omics characteristics and the filtered dose omics characteristics into multi-modal omics characteristics and inputting the multi-modal omics characteristics into different classifiers; and extracting multi-modal omics characteristics with optimal performance in each class, and establishing a radiation pneumonitis prediction model according to the multi-modal omics characteristics with optimal performance. Accurate identification and segmentation of clinical target area CTV are realized, automatic diagnosis, positioning and filtering analysis of lesion features of LUNG tissue LUNG are facilitated, intelligent identification of radiation pneumonitis and disease prediction are realized, judgment efficiency of radiation pneumonitis is improved, and medical safety of critical patients is guaranteed.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present application.
It should be understood that the above-described embodiments are merely exemplary of some, and not all, embodiments of the present application, and that the drawings illustrate preferred embodiments of the present application without limiting the scope of the claims appended hereto. This application is capable of embodiments in many different forms and is provided for the purpose of enabling a thorough understanding of the disclosure of the application. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to one skilled in the art that modifications can be made to the embodiments described in the foregoing detailed description, or equivalents can be substituted for some of the features described therein. All equivalent structures made by using the contents of the specification and the drawings of the present application are directly or indirectly applied to other related technical fields and are within the protection scope of the present application.
Claims (10)
1. A method for establishing a prediction model of radiation pneumonitis, which is characterized by comprising the following steps:
identifying LUNG tissue LUNG and clinical target area CTV in the CT image, and segmenting the LUNG and the CTV by using an image segmentation algorithm to obtain LUNG tissue LUNG-CTV after the clinical target area CTV is removed;
obtaining three-dimensional physical dose of LUNG tissue Lung-CTV region in radiotherapy plan of LUNG cancer patient, and passing through EQD 2 Converting the three-dimensional physical dose into 2Gy fractionated radiation equivalent biological dose by a formula;
extracting the imaging omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of 2Gy fractionated radiation equivalent dose;
carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm, and filtering the features representing redundant information;
combining the filtered image omics characteristics and the filtered dose omics characteristics into multi-modal omics characteristics and inputting the multi-modal omics characteristics into different classifiers;
and extracting multi-modal omics characteristics with optimal performance in each class, and establishing a radiation pneumonitis prediction model according to the multi-modal omics characteristics with optimal performance.
2. The method for building a prediction model of radiation pneumonitis according to claim 1, wherein the step of identifying LUNG tissue Lung and clinical target region CTV in CT image, and segmenting Lung and CTV to obtain LUNG tissue Lung-CTV after removing CTV comprises:
identifying LUNG tissue LUNG and clinical target region CTV in CT image;
calculating a marginal optimal threshold of LUNG tissue LUNG and a marginal optimal threshold of clinical target area CTV;
determining a contour edge of the LUNG tissue LUNG and a contour edge of the clinical target area CTV in the CT image based on the edge optimal threshold of the LUNG tissue LUNG and the edge optimal threshold of the clinical target area CTV;
segmenting LUNG tissue LUNG and clinical target region CTV, and obtaining LUNG tissue LUNG-CTV after removing clinical target region CTV.
3. The method for building a predictive model of radiation pneumonitis according to claim 2, wherein after the step of identifying LUNG tissue Lung and clinical target CTV in CT images, and segmenting Lung and CTV, and obtaining LUNG tissue Lung-CTV after removing clinical target CTV, the method comprises:
identifying a clinical target region CTV in LUNG tissue LUNG-CTV;
when no clinical target area CTV exists in the identified LUNG tissue LUNG-CTV, optimizing the edge of the LUNG tissue LUNG-CTV through mathematical morphology;
when the clinical target volume CTV is identified in the LUNG tissue LUNG-CTV, the step of calculating a marginal optimal threshold value of the LUNG tissue LUNG and a marginal optimal threshold value of the clinical target volume CTV is performed.
4. The method for building the prediction model of radiation pneumonitis as claimed in claim 1, wherein the three-dimensional physical dose of LUNG tissue Lung-CTV region in the radiation therapy planning of LUNG cancer patient is obtained by EQD 2 The step of converting the three-dimensional physical dose into the 2Gy fractionated radiation equivalent biological dose by the formula comprises the following steps:
acquiring a radiation therapy dose file of a lung cancer patient from a planning system;
extracting three-dimensional physical dose of LUNG tissue LUNG-CTV region in a LUNG cancer patient radiotherapy plan based on Python programming;
using EQD 2 The formula converts the three-dimensional physical dose into a 2Gy fractionated radiation equivalent biological dose.
5. The method for establishing the prediction model of the radiation pneumonitis according to claim 1, wherein the step of extracting the imaging omics characteristics of the LUNG tissue Lung-CTV region and the dosimetric characteristics after 2Gy fractionated radiation equivalent dose comprises:
extracting the image omics characteristics of the LUNG tissue LUNG-CTV region based on Python programming, wherein the image omics characteristics comprise at least one of tumor shape, gray distribution, texture characteristics, wavelet passing and Fourier transform;
the method comprises the steps of extracting dose omics characteristics in 2Gy fractionated radiation equivalent dose based on Python programming, wherein the dose omics characteristics comprise at least one of first-order statistics, 2D and 3D-based shape characteristics, gray level co-occurrence matrixes, gray level running length matrixes, gray level region size matrixes, gray level dependency matrixes, neighborhood gray level difference matrixes and the like.
6. The method for building a prediction model of radiation pneumonitis according to claim 1, wherein the step of performing feature screening on the iconomics features and the dosimetomics features by using a random forest algorithm and filtering the features representing redundant information comprises:
carrying out feature screening on the image omics features and the dose omics features by using a random forest algorithm;
acquiring the characteristics representing the redundant information and the characteristics representing the vector information;
filtering the characteristics of the characterization redundant information, and sequentially arranging the characteristics of the plurality of characterization vector information;
and acquiring the characteristics of the characterization vector information higher than the preset characterization vector, and setting the characteristics as multimodality omic characteristics.
7. The method for building the prediction model of radiation pneumonitis according to claim 6, wherein the step of combining the filtered imagery omics features and dosimetomics features into multimodality omics features and inputting the multimodality characteristics into different classifiers comprises:
classifying multimodality omics features based on a feature selection algorithm;
and (4) performing iterative learning on the multimodality omic features by using a classifier, and extracting the multimodality omic features with the optimal performance in each class.
8. A system for building a prediction model of radiation pneumonitis, the system comprising:
the identification module is used for identifying LUNG tissue LUNG and a clinical target area CTV in the CT image, segmenting the LUNG and the CTV and obtaining LUNG tissue LUNG-CTV after the clinical target area CTV is removed;
a transformation module for obtaining three-dimensional physical dose of LUNG tissue Lung-CTV region in radiotherapy planning of LUNG cancer patient, and passing through EQD 2 Converting the three-dimensional physical dose into 2Gy fractionated radiation equivalent biological dose by a formula;
the grading module is used for extracting the imaging omics characteristics of the Lung tissue LUNG-CTV region and the dose omics characteristics of 2Gy graded radiation equivalent dose;
the filtering module is used for performing feature screening on the image omics features and the dose omics features by using a random forest algorithm and filtering the features representing redundant information;
the classification module is used for forming multi-modal omics characteristics by the filtered image omics characteristics and the filtered dose omics characteristics and inputting the multi-modal omics characteristics into different classifiers;
and the establishing module is used for extracting the multimodality omics characteristics with optimal performance in each class and establishing the radiation pneumonitis prediction model according to the multimodality characteristics with optimal performance.
9. A computer device comprising a memory having computer readable instructions stored therein and a processor that when executed performs the steps of the method of building a radiation pneumonitis prediction model according to any one of claims 1 to 7.
10. A computer-readable storage medium, having computer-readable instructions stored thereon, which, when executed by a processor, implement the steps of the method for building a radiation pneumonitis prediction model according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210765929.8A CN115100155B (en) | 2022-07-01 | 2022-07-01 | Method and system for establishing radiation pneumonitis prediction model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210765929.8A CN115100155B (en) | 2022-07-01 | 2022-07-01 | Method and system for establishing radiation pneumonitis prediction model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115100155A true CN115100155A (en) | 2022-09-23 |
| CN115100155B CN115100155B (en) | 2024-08-23 |
Family
ID=83294419
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210765929.8A Active CN115100155B (en) | 2022-07-01 | 2022-07-01 | Method and system for establishing radiation pneumonitis prediction model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115100155B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117877737A (en) * | 2024-03-12 | 2024-04-12 | 北方健康医疗大数据科技有限公司 | Method, system and device for constructing primary lung cancer risk prediction model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110604550A (en) * | 2019-09-24 | 2019-12-24 | 广州医科大学附属肿瘤医院 | Prediction method of normal tissue organ complications after tumor radiotherapy |
| US20210110540A1 (en) * | 2019-10-11 | 2021-04-15 | Case Western Reserve University | Predicting tumor prognoses based on a combination of radiomic and clinico-pathological features |
| 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 |
| CN114266729A (en) * | 2021-11-29 | 2022-04-01 | 厦门大学附属第一医院 | Chest tumor radiotherapy-based radiation pneumonitis prediction method and system based on machine learning |
| CN114496239A (en) * | 2021-12-03 | 2022-05-13 | 汕头大学医学院附属肿瘤医院 | Novel lung cancer SBRT (systemic lupus erythematosus) radiation pneumonia risk prediction factor and construction method thereof |
-
2022
- 2022-07-01 CN CN202210765929.8A patent/CN115100155B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110604550A (en) * | 2019-09-24 | 2019-12-24 | 广州医科大学附属肿瘤医院 | Prediction method of normal tissue organ complications after tumor radiotherapy |
| US20210110540A1 (en) * | 2019-10-11 | 2021-04-15 | Case Western Reserve University | Predicting tumor prognoses based on a combination of radiomic and clinico-pathological features |
| 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 |
| CN114266729A (en) * | 2021-11-29 | 2022-04-01 | 厦门大学附属第一医院 | Chest tumor radiotherapy-based radiation pneumonitis prediction method and system based on machine learning |
| CN114496239A (en) * | 2021-12-03 | 2022-05-13 | 汕头大学医学院附属肿瘤医院 | Novel lung cancer SBRT (systemic lupus erythematosus) radiation pneumonia risk prediction factor and construction method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 陈文韬: "基于机器学习的肺癌放疗放射性肺炎预测研究", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》, 15 March 2022 (2022-03-15), pages 4 - 2 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117877737A (en) * | 2024-03-12 | 2024-04-12 | 北方健康医疗大数据科技有限公司 | Method, system and device for constructing primary lung cancer risk prediction model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115100155B (en) | 2024-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112270666A (en) | Non-small cell lung cancer pathological section identification method based on deep convolutional neural network | |
| Kaur et al. | Feature extraction and principal component analysis for lung cancer detection in CT scan images | |
| Huang et al. | Optimal deep learning neural network using ISSA for diagnosing the oral cancer | |
| Junayed et al. | ScarNet: development and validation of a novel deep CNN model for acne scar classification with a new dataset | |
| Dobay et al. | Potential use of deep learning techniques for postmortem imaging | |
| CN101609485A (en) | Medical image diagnostic system and diagnostic method based on the migration nuclear matching tracing | |
| Das et al. | Fusion of handcrafted and deep convolutional neural network features for effective identification of diabetic foot ulcer | |
| Abdollahi | Evaluating LeNet Algorithms in Classification Lung Cancer from Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases | |
| Tian et al. | Radiomics and its clinical application: artificial intelligence and medical big data | |
| CN115564756A (en) | Medical image focus positioning display method and system | |
| Arif et al. | Application of Genetic Algorithm and U‐Net in Brain Tumor Segmentation and Classification: A Deep Learning Approach | |
| CN115100155B (en) | Method and system for establishing radiation pneumonitis prediction model | |
| Oota et al. | WSNet: towards an effective method for wound image segmentation | |
| Khan | Identification of lung cancer using convolutional neural networks based classification | |
| Xu et al. | Identification of benign and malignant lung nodules in CT images based on ensemble learning method | |
| Nair et al. | Prediction and Classification of CT images for Early Detection of Lung Cancer Using Various Segmentation Models | |
| Al-Haija et al. | Development of breast cancer detection model using transfer learning of residual neural network (resnet-50) | |
| Wankhade et al. | Lung cell cancer identification mechanism using deep learning approach | |
| Anas et al. | Advancing Breast Cancer Detection: Enhancing YOLOv5 Network for Accurate Classification in Mammogram Images | |
| Ma et al. | Automatic pulmonary ground‐glass opacity nodules detection and classification based on 3D neural network | |
| Kukreti et al. | Detection and Classification of Brain Tumour Using EfficientNet and Transfer Learning Techniques | |
| Sorour et al. | An Automatic Dermatology Detection System Based on Deep Learning and Computer Vision | |
| Boumaraf et al. | Conventional Machine Learning versus Deep Learning for Magnification Dependent Histopathological Breast Cancer Image Classification: A Comparative Study with Visual Explanation. Diagnostics, 2021; 11 (3): 528 | |
| Vimala et al. | Novel Brain Tumor Classification Model with MLPNN Using UNET | |
| Essaf et al. | Review on deep learning methods used for computer-aided lung cancer detection and diagnosis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |