CN108159576A - Human body chest and abdomen surface region respiratory movement predicting method in a kind of radiotherapy - Google Patents

Human body chest and abdomen surface region respiratory movement predicting method in a kind of radiotherapy Download PDF

Info

Publication number
CN108159576A
CN108159576A CN201711359660.9A CN201711359660A CN108159576A CN 108159576 A CN108159576 A CN 108159576A CN 201711359660 A CN201711359660 A CN 201711359660A CN 108159576 A CN108159576 A CN 108159576A
Authority
CN
China
Prior art keywords
respiratory movement
radiotherapy
human body
hyper parameter
boundary line
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
Application number
CN201711359660.9A
Other languages
Chinese (zh)
Other versions
CN108159576B (en
Inventor
于晓洋
史领
韩玉翠
赵烟桥
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin University of Science and Technology
Original Assignee
Harbin University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Harbin University of Science and Technology filed Critical Harbin University of Science and Technology
Priority to CN201711359660.9A priority Critical patent/CN108159576B/en
Publication of CN108159576A publication Critical patent/CN108159576A/en
Application granted granted Critical
Publication of CN108159576B publication Critical patent/CN108159576B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • 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
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G06T7/248—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments involving reference images or patches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
    • 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
    • A61N2005/1059—Monitoring, verifying, controlling systems and methods for verifying the position of the patient with respect to the radiation beam using cameras imaging the patient
    • 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
    • A61N2005/1092—Details
    • 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
    • G06T2207/10016—Video; Image sequence
    • 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
    • G06T2207/10024—Color image
    • 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)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Veterinary Medicine (AREA)
  • Radiology & Medical Imaging (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Pathology (AREA)
  • Public Health (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Radiation-Therapy Devices (AREA)

Abstract

Human body chest and abdomen surface region respiratory movement predicting method belongs to oncotherapy and respiratory movement prediction field in a kind of radiotherapy of the present invention；The Forecasting Methodology with respiratory movement forecast function acquires respiratory movement signal data using twin camera, and the data after acquisition are pre-processed, using obtained data as the input of Gaussian process regression forecasting algorithm.Then propose that one kind predicts regression problem based on Gaussian process model.A suitable regression model is built, and be trained and predict using data, select kernel function in the training process, and hyper parameter therein is obtained, pass through off-line simulation experimental verification algorithm feasibility.Hyper parameter is obtained using conjugate gradient method in finally selection kernel function paracycle, carries out Gaussian process homing method and carries out model prediction to respiratory movement data.Human body chest and abdomen surface region respiratory movement predicting method in radiotherapy of the present invention can reduce respiratory movement for influence caused by radiotherapy precision.

Description

Human body chest and abdomen surface region respiratory movement predicting method in a kind of radiotherapy
Technical field
Human body chest and abdomen surface region respiratory movement predicting method belongs to respiratory movement electric powder prediction in a kind of radiotherapy.
Background technology
Respiratory movement can be such that tumour and normal structure is moved with certain frequency and amplitude, this causes practical in Radiotherapy Difference between absorbed dose of radiation and planning system result of calculation simultaneously influences therapeutic effect or even causes the radiation-induced damage of human body.Radiation Respiratory movement is the main reason for causing breast belly tumour that position movement and volume change occur therewith, as a result to add in therapeutic process The phenomenon that leakage of weight target area is according to normal structure by according to increasing, and respiratory movement causes neoplastic transformation and movement and leads to practical suction Dosage is received with planning the difference of metering room, this influence to dosage effect is clinically very significant.
Conventional at present includes for handling respirometric method in radiotherapy：Movement comprising method, compression type shallow breathing method, The clinical technologies such as method of holding one's breath and respiration gate control method, produce positive effect, but because of patient tolerance to thorax and abdomen malignant radiotherapy Property is poor, therapeutic efficiency is low, normal structure is still by compared with reasons such as major injuries, also reaches desired effect far away.
The algorithm of the respiratory movement prediction of implementation has two classes.The first kind, the prediction mould based on mathematics and physical method Type；Second class, the prediction model based on modern science and technology and method.The former includes linear estimation algorithm and time series ARIMA (Autoregressive Integrated Moving Average Model) in model, this is most widely used A kind of general temporal model, subsequence matching model, Kalman filter model (Kalman Filtering Model), parameter are returned Return model (Parametric Regressive Model), exponential smoothing model (Exponential Smoothing Model various combination forecastings) and by these models formed etc..And the latter has nonparametric Regression Model (Nonparametric Regressive Model), KARIMA algorithms, adaptive weighting model, spectrum analysis method (Spectral Basis Analysis), I reconstruction models of state sky, wavelet network (Wavelet Network), the side of shape is divided based on multidimensional Method, the method based on wavelet function feedback and a variety of and neural network (Neural Network) relevant hybrid model for short-term load forecasting Deng.
But compared with these models, the advantage of Gaussian process is first, can be by priori in the process with elder generation The form for testing probability represents, improves the performance of model.Second, Unknown worm item can be made with the defeated of precision parameter Go out prediction, which generally refers to estimate variance model parameter, is significantly reduced by parameter, the opposite appearance of parameter optimization Easily, easily the character representations such as convergence come out parameter.And by the continuous exploration of nearly 10 years domestic and foreign scholars, Gaussian process exists Approved in practice, significantly improved in supervised study using degree.
Invention content
To solve the above-mentioned problems, the present invention provides human body chest and abdomen surface region respiratory movement prediction sides in a kind of radiotherapy Method is somebody's turn to do the respiratory movement forecast analysis based on Gaussian process, by building correlation models and prediction model and combining the two Finally one more accurate Forecasting Methodology of structure, and then provided more for tumour respiratory movement during monitoring and predicting radiotherapy Good basis, solves the problems, such as by respiratory movement to be caused in Patients During Radiotherapy.
The object of the present invention is achieved like this：
Human body chest and abdomen surface region respiratory movement predicting method, includes the following steps in a kind of radiotherapy：
Step a, the width multicolour pattern that three different frequency RGB cosine curves are combined using the light pattern projector Human body chest and abdomen surface is projected, light pattern projector both sides are respectively placed a 3CCD colour TV cameras acquisition scene image and are sent into Computer is post-processed, and obtains the three-dimensional in feature mark poiX and regional edge boundary line according to Binocular Vision Principle using twin camera Coordinate；
Step b, for left and right cameras obtain two video sequences, extracted in correspondence image pair the same area and Behind its boundary line and feature mark poiX, difference matching area boundary line and feature mark poiX, then using left and right cameras foundation Binocular Vision Principle is obtained the three-dimensional coordinate of point and feature mark poiX on boundary line, is combined using video camera with the projector according to item Line is analyzed and phase developing method obtains the three-dimensional coordinate of region inner surface point, and characteristic point can be calculated further according to mathematical definition Three-dimensional coordinate, regional edge boundary line floor projection perimeter and geometric center, regional edge boundary line each point coordinate average value and perimeter, region Each point coordinates average value and surface area amount to 7 provincial characteristics amounts；
Step c, according to specific predicted characteristics amount present position, area-of-interest and its boundary and feature mark poiX are determined； Using predicted characteristics amount training observation value as reference, carry out correlation analysis with other all areas characteristic quantity training observation values and show The analysis of work property, optimization carries out participating in modeling and the provincial characteristics duration set of prediction is denoted as Y；
Step d, selection kernel function paracycle is as follows
Wherein, r=| | x-x'| |2Represent the Euclidean distance between two data points, θS、θL、θpFor hyper parameter；
Step e, to ensure KcFor effective positive definite covariance function, using Cholesky factorization and to lower triangular matrix Element parameterized, obtain Kc=L (θc)L(θc)T, wherein L (θc) it is a lower triangular matrix, size is m × m；
Step f, i.e.-log (y | θ) is minimized about hyper parameter for negative logarithm marginal probability, then using conjugation Gradient method asks for its optimal value；
Step g, by the cross-correlation between estimation range characteristic quantity and other provincial characteristics amounts maximized come to Mobile hyper parameter θSInitial value gives hyper parameter θ according to the training data of other provincial characteristics amountspInitial value is assigned, repeats to test repeatedly, His hyper parameter random initializtion；
Step h, in measurement and forecast period, other unexpected areas of estimation range characteristic quantity are measured first, in accordance with sample frequency Characteristic of field amount, then for prediction time x*=t+ Δs t predicted and provides predictive estimation value and its error confidence interval, Middle Δ t is predicted time.
Human body chest and abdomen surface region respiratory movement prediction model in above-mentioned radiotherapy, the L (θc) nonzero element by θcCome Regulation, related hyper parameter θcNumber be
Human body chest and abdomen surface region respiratory movement prediction model in above-mentioned radiotherapy, forming the sufficient and necessary condition of kernel function is The matrix formed between test centrostigma and point must be positive semidefinite matrix.
Advantageous effect：Accurately prediction tumour is heavily dependent on perfect real-time tracking forecasting system, the system It can help doctor's real-time tracking tumor motion state and predict the position at tumour lower a moment in time, contribute to radiation beam again Adjustment, avoids the injury to health tissues, this is the important method for improving oncotherapy curative effect.The prediction algorithm pair of pinpoint accuracy The readjusting of radiation beam plays the role of vital, can ensure that radiation beam kills the mistake of tumor carcinoma cells completely Cheng Zhong, normal tissue injury is minimum, this is that very have clinical meaning.
By domestic and international comparative analysis, it is found that mathematically Gaussian process and large-scale neural network, Bayesian model etc. are permitted The equivalence more being known.But compared with these models, the Gaussian process regressive prediction model that is used in the present invention Advantage is first, priori in the process can be represented in the form of prior probability, improve the performance of model.The Two, the output prediction with precision parameter can be made to Unknown worm item, which generally refers to estimate variance Model parameter, significantly reduced by parameter, parameter optimization it is relatively easy, parameter be easier restrain etc. character representations come out.And By the continuous exploration of nearly 10 years domestic and foreign scholars, Gaussian process is approved in practice, is applied in supervised study Degree significantly improves.
Description of the drawings
Fig. 1 is Gaussian process regression forecasting result.
Specific embodiment
The specific embodiment of the invention is described in further detail below in conjunction with the accompanying drawings.
Specific embodiment one
Human body chest and abdomen surface region respiratory movement predicting method, includes the following steps in a kind of radiotherapy of the present embodiment：
Step a, the width multicolour pattern that three different frequency RGB cosine curves are combined using the light pattern projector Human body chest and abdomen surface is projected, light pattern projector both sides are respectively placed a 3CCD colour TV cameras acquisition scene image and are sent into Computer is post-processed, and obtains the three-dimensional in feature mark poiX and regional edge boundary line according to Binocular Vision Principle using twin camera Coordinate；
Step b, for left and right cameras obtain two video sequences, extracted in correspondence image pair the same area and Behind its boundary line and feature mark poiX, difference matching area boundary line and feature mark poiX, then using left and right cameras foundation Binocular Vision Principle is obtained the three-dimensional coordinate of point and feature mark poiX on boundary line, is combined using video camera with the projector according to item Line is analyzed and phase developing method obtains the three-dimensional coordinate of region inner surface point, and characteristic point can be calculated further according to mathematical definition Three-dimensional coordinate, regional edge boundary line floor projection perimeter and geometric center, regional edge boundary line each point coordinate average value and perimeter, region Each point coordinates average value and surface area amount to 7 provincial characteristics amounts；
Step c, according to specific predicted characteristics amount present position, area-of-interest and its boundary and feature mark poiX are determined； Using predicted characteristics amount training observation value as reference, carry out correlation analysis with other all areas characteristic quantity training observation values and show The analysis of work property, optimization carries out participating in modeling and the provincial characteristics duration set of prediction is denoted as Y；
Step d, selection kernel function paracycle is as follows
Wherein, r=| | x-x'| |2Represent the Euclidean distance between two data points, θ S, θ L, θpFor hyper parameter；Structure Sufficient and necessary condition into kernel function is that the matrix tested centrostigma and formed between putting must be positive semidefinite matrix；
Step e, to ensure KcFor effective positive definite covariance function, using Cholesky factorization and to lower triangular matrix Element parameterized, obtain Kc=L (θc)L(θc)T, wherein L (θc) it is a lower triangular matrix, size is m × m；L(θc) Nonzero element by θcIt provides, related hyper parameter θcNumber be
Step f, i.e.-log (y | θ) is minimized about hyper parameter for negative logarithm marginal probability, then using conjugation Gradient method asks for its optimal value；
Step g, by the cross-correlation between estimation range characteristic quantity and other provincial characteristics amounts maximized come to Mobile hyper parameter θSInitial value gives hyper parameter θ according to the training data of other provincial characteristics amountspInitial value is assigned, repeats to test repeatedly, His hyper parameter random initializtion；
Step h, in measurement and forecast period, other unexpected areas of estimation range characteristic quantity are measured first, in accordance with sample frequency Then characteristic of field amount predicted for prediction time x*=t+ Δ t and provides predictive estimation value and its error confidence interval, Middle Δ t is predicted time.
Finally obtain Gaussian process regression forecasting as shown in Figure 1 as a result, wherein dark line be raw data plot, light line For prediction result curve, the region of grey represents its confidence interval.

Claims (3)

1. human body chest and abdomen surface region respiratory movement predicting method in a kind of radiotherapy, which is characterized in that include the following steps：
Step a, it is projected using the width multicolour pattern that three different frequency RGB cosine curves are combined by the light pattern projector To human body chest and abdomen surface, light pattern projector both sides respectively place a 3CCD colour TV cameras acquisition scene image and are sent into calculating Machine is post-processed, and obtains feature mark poiX according to Binocular Vision Principle using twin camera and the three-dimensional of regional edge boundary line is sat Mark；
Step b, two video sequences obtained for left and right cameras, extract the same area and its side in correspondence image pair Behind boundary line and feature mark poiX, difference matching area boundary line and feature mark poiX, then using left and right cameras according to binocular Visual theory is obtained the three-dimensional coordinate of point and feature mark poiX on boundary line, is combined using video camera with the projector according to striped minute Analysis and phase developing method obtain the three-dimensional coordinate of region inner surface point, and characteristic point three-dimensional can be calculated further according to mathematical definition Coordinate, regional edge boundary line floor projection perimeter and geometric center, regional edge boundary line each point coordinate average value and perimeter, region each point Coordinate average value and surface area amount to 7 provincial characteristics amounts；
Step c, according to specific predicted characteristics amount present position, area-of-interest and its boundary and feature mark poiX are determined；With pre- Characteristic quantity training observation value is surveyed as reference, correlation analysis and conspicuousness are carried out with other all areas characteristic quantity training observation values Analysis, optimization carries out participating in modeling and the provincial characteristics duration set of prediction is denoted as Y；
Step d, selection kernel function paracycle is as follows
Wherein, r=| | x-x'| |2Represent the Euclidean distance between two data points, θS、θL、θpFor hyper parameter；
Step e, to ensure KcFor effective positive definite covariance function, using Cholesky factorization and to the member of lower triangular matrix Element is parameterized, and obtains Kc=L (θc)L(θc)T, wherein L (θc) it is a lower triangular matrix, size is m × m；
Step f, i.e.-log (y | θ) is minimized about hyper parameter for negative logarithm marginal probability, then using conjugate gradient Method asks for its optimal value；
Step g, it is maximized by the cross-correlation between estimation range characteristic quantity and other provincial characteristics amounts come to movement Hyper parameter θSInitial value is assigned, hyper parameter θ is given according to the training data of other provincial characteristics amountspInitial value is assigned, repeats to test repeatedly, other Hyper parameter random initializtion；
Step h, in measurement and forecast period, it is special that other unexpected regions of estimation range characteristic quantity are measured first, in accordance with sample frequency Sign amount, then for prediction time x*=t+ Δs t predicted and provides predictive estimation value and its error confidence interval, wherein Δ T is predicted time.
2. human body chest and abdomen surface region respiratory movement predicting method in radiotherapy according to claim 1, it is characterised in that：Institute L (the θ statedc) nonzero element by θcIt provides, related hyper parameter θcNumber be
3. human body chest and abdomen surface region respiratory movement predicting method in radiotherapy according to claim 1, it is characterised in that：Structure Sufficient and necessary condition into kernel function is that the matrix tested centrostigma and formed between putting must be positive semidefinite matrix.
CN201711359660.9A 2017-12-17 2017-12-17 Human body chest and abdomen surface area respiratory motion prediction method in radiotherapy Expired - Fee Related CN108159576B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711359660.9A CN108159576B (en) 2017-12-17 2017-12-17 Human body chest and abdomen surface area respiratory motion prediction method in radiotherapy

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711359660.9A CN108159576B (en) 2017-12-17 2017-12-17 Human body chest and abdomen surface area respiratory motion prediction method in radiotherapy

Publications (2)

Publication Number Publication Date
CN108159576A true CN108159576A (en) 2018-06-15
CN108159576B CN108159576B (en) 2020-01-07

Family

ID=62522113

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711359660.9A Expired - Fee Related CN108159576B (en) 2017-12-17 2017-12-17 Human body chest and abdomen surface area respiratory motion prediction method in radiotherapy

Country Status (1)

Country Link
CN (1) CN108159576B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109741831A (en) * 2019-01-09 2019-05-10 哈尔滨理工大学 The human body chest and abdomen surface region breath signal period forecasting method calculated based on variance
CN109741828A (en) * 2019-01-09 2019-05-10 哈尔滨理工大学 A kind of human body chest and abdomen surface region breath signal period forecasting method in radiotherapy
CN110681074A (en) * 2019-10-29 2020-01-14 苏州大学 Tumor respiratory motion prediction method based on bidirectional GRU network
CN113674393A (en) * 2021-07-12 2021-11-19 中国科学院深圳先进技术研究院 Construction method of respiratory motion model and unmarked respiratory motion prediction method
CN114177545A (en) * 2022-01-17 2022-03-15 中国科学院合肥物质科学研究院 Non-contact respiratory rhythm monitoring device and method used in radiotherapy
CN115187608A (en) * 2022-09-14 2022-10-14 苏州大学 Respiration characteristic extraction method based on body surface significance analysis

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103761745A (en) * 2013-07-31 2014-04-30 深圳大学 Estimation method and system for lung motion model
CN104984481A (en) * 2015-05-20 2015-10-21 清华大学 Method and system for carrying out on-line compensation on tumor motion generated by breathing

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103761745A (en) * 2013-07-31 2014-04-30 深圳大学 Estimation method and system for lung motion model
CN104984481A (en) * 2015-05-20 2015-10-21 清华大学 Method and system for carrying out on-line compensation on tumor motion generated by breathing

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
万伟权: "《放疗中肿瘤运动基于实时跟踪呼吸预测的算法》", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 *
欧阳斌: "《基于体外信号的呼吸运动跟踪模型的研究》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109741831A (en) * 2019-01-09 2019-05-10 哈尔滨理工大学 The human body chest and abdomen surface region breath signal period forecasting method calculated based on variance
CN109741828A (en) * 2019-01-09 2019-05-10 哈尔滨理工大学 A kind of human body chest and abdomen surface region breath signal period forecasting method in radiotherapy
CN109741828B (en) * 2019-01-09 2022-11-25 哈尔滨理工大学 Human chest and abdomen surface area respiratory signal period prediction method in radiotherapy
CN109741831B (en) * 2019-01-09 2022-12-06 哈尔滨理工大学 Human chest and abdomen surface area respiratory signal period prediction method based on variance calculation
CN110681074A (en) * 2019-10-29 2020-01-14 苏州大学 Tumor respiratory motion prediction method based on bidirectional GRU network
CN110681074B (en) * 2019-10-29 2021-06-15 苏州大学 Tumor respiratory motion prediction method based on bidirectional GRU network
CN113674393A (en) * 2021-07-12 2021-11-19 中国科学院深圳先进技术研究院 Construction method of respiratory motion model and unmarked respiratory motion prediction method
CN113674393B (en) * 2021-07-12 2023-09-26 中国科学院深圳先进技术研究院 Method for constructing respiratory motion model and method for predicting unmarked respiratory motion
CN114177545A (en) * 2022-01-17 2022-03-15 中国科学院合肥物质科学研究院 Non-contact respiratory rhythm monitoring device and method used in radiotherapy
CN114177545B (en) * 2022-01-17 2023-11-07 中国科学院合肥物质科学研究院 Contactless respiratory rhythm monitoring device and method for radiotherapy
CN115187608A (en) * 2022-09-14 2022-10-14 苏州大学 Respiration characteristic extraction method based on body surface significance analysis

Also Published As

Publication number Publication date
CN108159576B (en) 2020-01-07

Similar Documents

Publication Publication Date Title
CN108159576A (en) Human body chest and abdomen surface region respiratory movement predicting method in a kind of radiotherapy
CN110378881B (en) Tumor positioning system based on deep learning
TW201249405A (en) System for facilitating operation of treatment delivery system and method for controlling operation of treatment delivery system
Sarrut et al. A comparison framework for breathing motion estimation methods from 4-D imaging
CN108154503A (en) A kind of leucoderma state of an illness diagnostic system based on image procossing
Foote et al. Real-time 2D-3D deformable registration with deep learning and application to lung radiotherapy targeting
US20130346050A1 (en) Method and apparatus for determining focus of high-intensity focused ultrasound
WO2021042807A1 (en) Method for designing bilateral filtering-based motion compensation high-quality 4d-cbct image reconstruction
CN107067393A (en) A kind of three-dimensional medical image segmentation method based on user mutual and shape prior knowledge
Zhang et al. A kernel-based method for markerless tumor tracking in kV fluoroscopic images
CN103345741A (en) Non-rigid multimode medical image precise registering method
CN113499091B (en) Method and system for predicting tumor movement correlation and tumor internal mobility in body surface and body of patient
Aghajani et al. A robust image registration method based on total variation regularization under complex illumination changes
de Souza et al. Predicting body measures from 2D images using Convolutional Neural Networks
Zhang et al. Dynamic lung modeling and tumor tracking using deformable image registration and geometric smoothing
Foote et al. Real-time patient-specific lung radiotherapy targeting using deep learning
CN115187608B (en) Respiration characteristic extraction method based on body surface significance analysis
Gamal et al. Brain tumor segmentation using 3D U-Net with hyperparameter optimization
Romaguera et al. Conditional-based Transformer network with learnable queries for 4D deformation forecasting and tracking
Torshabi et al. A clinical application of fuzzy logic
Wang et al. An improved correlation model for respiration tracking in robotic radiosurgery using essential skin surface motion
Fu et al. Divergence-free fitting-based incompressible deformation quantification of liver
Abhilash et al. Empirical modeling of renal motion for improved targeting during focused ultrasound surgery
CN104268878B (en) Probe in detecting and localization method are moved in a kind of ultrasonoscopy
You et al. A novel application of principal surfaces to segmentation in 4D-CT for radiation treatment planning

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
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20200107

Termination date: 20201217

CF01 Termination of patent right due to non-payment of annual fee