CN114937068A - Method, system and device for obtaining tissue equivalent scintillator dose distribution - Google Patents
Method, system and device for obtaining tissue equivalent scintillator dose distribution Download PDFInfo
- Publication number
- CN114937068A CN114937068A CN202210378351.0A CN202210378351A CN114937068A CN 114937068 A CN114937068 A CN 114937068A CN 202210378351 A CN202210378351 A CN 202210378351A CN 114937068 A CN114937068 A CN 114937068A
- Authority
- CN
- China
- Prior art keywords
- scintillator
- tissue equivalent
- dose distribution
- image
- data set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/047—Probabilistic or stochastic networks
-
- 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
- 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]
Abstract
The invention discloses a method, a system and a device for obtaining tissue equivalent scintillator dose distribution, wherein the method comprises the following steps: s100, capturing scintillation light information generated by depositing energy in a tissue equivalent material through rays by a light field camera to form an image data set; s200, designing a convolutional neural network model based on the preprocessed image data set, and acquiring three-dimensional light information distribution of each image in the image data set by using an image depth estimation algorithm; s300, determining the dose distribution according to the acquired three-dimensional light information distribution. The invention designs a new method and a new device for acquiring the dose distribution, and the light information distribution determines the dose distribution, so that the three-dimensional dose distribution information of the rays in the tissue equivalent material can be more accurately measured.
Description
Technical Field
The invention relates to the technical field of dose measurement, in particular to a method, a system and a device for acquiring tissue equivalent scintillator dose distribution.
Background
Radiotherapy is widely applied to treatment of human diseases such as cancer, dosage estimation is crucial in the radiotherapy process, and accurate dosage estimation can avoid the risk of over-dosage irradiation of a patient. The light field imaging technology is used as a branch of modern image measurement and is widely applied to the fields of industrial manufacturing, machine vision and the like. And acquiring a light field image by using a light field imaging technology, performing inversion by using a depth estimation algorithm, further acquiring light field three-dimensional light distribution information of rays in the tissue equivalent scintillator, and finally accurately acquiring a dose distribution result.
The traditional image reconstruction algorithm is mainly based on the three-dimensional geometric properties of line segments in a light field image, the depth of the image is obtained by calculating parallax through line matching between sub-aperture images, and the problems of three-dimensional matching and low distance measurement precision exist. The image depth estimation method based on the correlation calculation does not need to calibrate a camera, and greatly simplifies the process of depth estimation. However, the baseline of the virtual camera between multiple viewing angles is too short, which may cause a problem of mismatching, and there is no clear proportional relationship between the correlation of the image block and the depth information, so that it is difficult to quantitatively analyze the result. Therefore, a new image reconstruction algorithm is needed to obtain the distribution of three-dimensional light information of the tissue equivalent scintillator.
Disclosure of Invention
In view of the drawbacks existing in the prior art, the present invention aims to provide a method, a system and a device for obtaining tissue equivalent scintillator dose distribution, which can accurately measure three-dimensional dose distribution information of radiation in a tissue equivalent scintillator by using only scintillator materials and an imaging unit.
In order to realize the purpose, the technical scheme adopted by the invention is as follows:
a method of obtaining a tissue equivalent scintillator dose distribution, comprising the steps of:
capturing scintillation light information generated by depositing energy in a tissue equivalent scintillator in a physical model through a light field camera to form an image data set;
preprocessing the light field image in the image data set, designing a convolutional neural network model based on the preprocessed image data set, and acquiring three-dimensional light information distribution of each image in the image data set by using an image depth estimation algorithm.
And determining the dose distribution according to the acquired three-dimensional light information distribution.
Further, a method of obtaining a tissue equivalent scintillator dose distribution as above, comprising:
when the tissue equivalent scintillator in the physical model is subjected to energy deposition of rays of a radiation source, luminous information is generated, and the light yield is in direct proportion to the energy deposition of the rays;
the light field camera collects luminescence information generated by energy deposition of the rays in the tissue equivalent scintillator, and reconstructs a light field image of three-dimensional emission light formed after the physical model is radiated according to the collected luminescence information to form an image data set.
Further, a method of obtaining a tissue equivalent scintillator dose distribution as above, comprising:
respectively extracting feature vectors in the horizontal direction and the vertical direction of the image in the image data set based on the preprocessed image data set and the convolutional neural network model;
and receiving the feature vector output by the convolutional neural network through a softmax function, and acquiring the three-dimensional light information of each image in the image data set.
The embodiment of the invention also provides a system for acquiring tissue equivalent scintillator dose distribution, which comprises the following modules:
the acquisition module is used for capturing scintillation light information generated by depositing energy in the tissue equivalent material through rays by a light field camera to form an image data set;
the calculation module is used for designing a convolutional neural network model based on the preprocessed image data set to obtain the three-dimensional light information distribution of each image in the image data set;
and the output module is used for determining the dose distribution according to the acquired three-dimensional light information distribution.
Further, the acquisition module reconstructs a light field image of three-dimensional emission light formed after the physical model is irradiated according to the collected luminescence information.
Further, the calculation module trains the convolutional neural network model according to the distribution characteristics of the light rays and the image characteristics in the horizontal direction and the vertical direction.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the aforementioned method of obtaining a tissue equivalent scintillator dose distribution.
An apparatus for obtaining a tissue equivalent scintillator dose distribution, comprising:
a processor; and
a memory for storing executable instructions of the processor;
wherein the processor is configured to perform the aforementioned method of obtaining a tissue equivalent scintillator dose distribution via execution of the executable instructions.
The invention has the beneficial effects that: according to the method, the system and the device, the light field imaging equipment is adopted, the convolutional neural network is designed to carry out light field image depth estimation, the high resolution of image acquisition and the accuracy of image depth information are met, and finally the dose distribution of rays in the tissue equivalent scintillator is reconstructed. Technical support can be provided for the field of scintillator-based three-dimensional dose distribution measurement.
Drawings
FIG. 1 is a schematic structural diagram of an apparatus for obtaining a tissue equivalent scintillator dose distribution according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for obtaining a tissue equivalent scintillator dose distribution provided in an embodiment of the present invention;
FIG. 3 is a flow chart of an image depth estimation method provided in an embodiment of the present invention;
fig. 4 is a block diagram of a device for implementing the method for obtaining a tissue equivalent scintillator dose in this embodiment.
Detailed Description
The invention is described in further detail below with reference to the drawings and the detailed description.
As shown in fig. 1, the camera takes a tissue equivalent scintillator dose real object conveyed by a conveyor belt and then processes photographic information stored in the camera using a computer.
As shown in fig. 2, the present embodiment provides a method for obtaining a tissue equivalent scintillator dose distribution, including:
s100, capturing scintillation light information generated by depositing energy in a tissue equivalent scintillator in a physical model through a light field camera to form an image data set;
specifically, S100 includes:
s101, when a scintillator in the physical model is subjected to energy deposition of rays of a radiation source, luminescence information is generated, and the light yield is in direct proportion to the energy deposition of the rays;
s102, the light field camera collects the luminescence information generated by the physical model, and reconstructs a light field image of three-dimensional emission light formed after the physical model is radiated according to the collected luminescence information to form an image data set.
The light field image is obtained by a light field camera with 4 x 4 microlens array, and the focal length of the light field camera is adjusted to enable the light field camera to focus accurately. And keeping the parameters of the camera unchanged, and transforming the scene to obtain more than 200 light field images. The resolution of each sub-image is 512 x 512.
S200, preprocessing the light field image in the image data set, designing a convolutional neural network model based on the preprocessed image data set, performing model training and feature learning, and acquiring three-dimensional light information of each image in the image data set by using an image depth estimation algorithm.
The specific image depth estimation algorithm is shown in fig. 3, and S200 includes:
s201, because the convolutional neural network is sensitive to whether the data set is balanced or not, a network model trained by an unbalanced data set is poor in classification effect. Therefore, it is necessary to balance the data, so that the data of different label classes are distributed close to the average, and the parallelism of the sample set, i.e. the number of training samples of different features, is kept substantially the same. The data set may then be scaled into 5 shares (keeping the class scale the same in each share), with 4 being the training set and 1 being the test set.
S202, determining the depth range of the light field image to be m, n according to the parameter distribution of the training data set]And the spacing of two adjacent fields is set to d, whereby the depth can be divided intoAnd (4) class. The EPI area blocks in the horizontal direction and the vertical direction are input into the sub-network as a pair of features, and each feature vector is output. Finally, feature vectors from the two sub-networks are received by utilizing a softmax function, the depth information of the images is predicted, and the three-dimensional light information of each image in the image data set is obtained.
And S300, determining the dose distribution according to the acquired three-dimensional light information distribution.
In this embodiment, the related training parameters of the convolutional neural network model are set as follows: the value of Batchsize (which refers to the number of samples selected for one training) is 64, the learning rate is set to 0.01, the learning rate reduction factor is set to 0.99, dropout is set to 0.5, and the number of iterations is 20.
The invention designs a novel method for acquiring tissue equivalent scintillator dose distribution, and model training and feature learning are carried out by taking a light field image as a data set, so that light distribution in a tissue equivalent scintillator is finally acquired. A three-dimensional dose measurement system formed by the tissue equivalent scintillating material and the light field camera can accurately measure the three-dimensional dose distribution information of rays in the tissue equivalent material, and provides more accurate reference for dose estimation in radiotherapy.
Meanwhile, referring to fig. 4, the embodiment further provides an apparatus for implementing the method for obtaining a tissue equivalent scintillator dosage, which includes:
the acquisition module 100 is used for capturing scintillation light information generated by depositing energy in a tissue equivalent material through rays by a light field camera to form an image data set;
a calculation module 200, configured to design a convolutional neural network model based on the preprocessed image dataset, and obtain three-dimensional light information distribution of each image in the image dataset by using an image depth estimation algorithm;
the output module 300 is configured to output a three-dimensional distribution of light, that is, determine a dose distribution according to the acquired three-dimensional light information distribution.
In this embodiment, a method and an apparatus for obtaining tissue equivalent scintillator dose distribution may be a computer, and include: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the aforementioned method of determining potentially important information for a patient via execution of the executable instructions. The memory and the processor may be connected by a bus. The memory unit may include a readable medium in the form of a volatile memory unit, such as a random access memory unit (RAM) and/or a cache memory unit, and may further include a read only memory unit (ROM). The computer also includes a display unit connected to the bus. The display unit may display the aforementioned important information and the like.
The present embodiment also provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by the processor, implements the aforementioned method for acquiring an optical information distribution.
In summary, the method for obtaining the tissue equivalent scintillator dose distribution and the apparatus for obtaining the tissue equivalent scintillator dose distribution in the present embodiment are the same inventive concept.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is intended to include such modifications and variations.
Claims (10)
1. A method of obtaining a tissue equivalent scintillator dose distribution, the method comprising the steps of:
s100, capturing scintillation light information generated by depositing energy in a tissue equivalent scintillator in a physical model through a light field camera to form an image data set;
s200, designing a convolutional neural network model based on the preprocessed image data set, and acquiring three-dimensional light information distribution of each image in the image data set by using an image depth estimation algorithm;
and S300, determining the dose distribution according to the acquired three-dimensional light information distribution.
2. The method for obtaining a tissue equivalent scintillator dose distribution as claimed in claim 1, wherein step S100 comprises:
s101, when the tissue equivalent scintillator in the physical model is subjected to energy deposition of rays of a radiation source, luminous information is generated, and the light yield is in direct proportion to the energy deposition of the rays;
s102, the light field camera collects luminescence information generated by deposition energy of the rays in the tissue equivalent scintillator, light field images of three-dimensional emission light formed after the physical model is radiated are reconstructed according to the collected luminescence information, and an image data set is formed.
3. The method of obtaining a tissue equivalent scintillator dose distribution as claimed in claim 1, wherein said light field camera comprises: the image sensor is aligned to the micro-lens array, the lens of the main lens is aligned to the physical model, and each micro-lens in the micro-lens array covers a plurality of sensor pixels.
4. The method for obtaining a tissue equivalent scintillator dose distribution as claimed in claim 1, wherein step S200 comprises:
s201, extracting horizontal and vertical polar line graphs corresponding to a plurality of pixel points from each light field image in the image data set to form a data set and a test set of model training;
s202, training a convolutional neural network model to obtain characteristic information of an image;
and S203, reconstructing a three-dimensional image to obtain three-dimensional distribution information of the ray in the tissue equivalent scintillator.
5. The method for obtaining the three-dimensional light information distribution of the tissue equivalent scintillator according to claim 4, wherein:
the convolutional neural network comprises two same sub-networks which are respectively used for training polar line graphs in the horizontal direction and the vertical direction and extracting characteristic vectors of the images in the horizontal direction and the vertical direction in the image data set.
6. A system for obtaining a tissue equivalent scintillator dose distribution, the system comprising:
the acquisition module is used for capturing scintillation light information generated by depositing energy in the tissue equivalent material through rays by a light field camera to form an image data set;
the calculation module is used for designing a convolutional neural network model based on the preprocessed image data set and acquiring three-dimensional light information distribution of each image in the image data set by using an image depth estimation algorithm;
and the output module is used for determining the dose distribution according to the acquired three-dimensional light information distribution.
7. The system for acquiring a tissue equivalent scintillator dose distribution as set forth in claim 6, wherein the acquisition module reconstructs a light field image of the three-dimensional emission light formed by the physical model after being irradiated based on the collected emission information.
8. The system for obtaining tissue equivalent scintillator dose distribution according to claim 6 or 7, wherein the calculation module trains the convolutional neural network model according to the image features in the horizontal and vertical directions according to the distribution features of the light.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method of obtaining a tissue equivalent scintillator dose distribution according to any one of claims 1 to 5.
10. An apparatus for obtaining a tissue equivalent scintillator dose distribution, comprising:
a processor; and
a memory for storing executable instructions of the processor;
wherein the processor is configured to perform the method of obtaining a tissue equivalent scintillator dose distribution of any of claims 1-5 via execution of the executable instructions.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210378351.0A CN114937068A (en) | 2022-04-12 | 2022-04-12 | Method, system and device for obtaining tissue equivalent scintillator dose distribution |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210378351.0A CN114937068A (en) | 2022-04-12 | 2022-04-12 | Method, system and device for obtaining tissue equivalent scintillator dose distribution |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114937068A true CN114937068A (en) | 2022-08-23 |
Family
ID=82862877
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210378351.0A Pending CN114937068A (en) | 2022-04-12 | 2022-04-12 | Method, system and device for obtaining tissue equivalent scintillator dose distribution |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114937068A (en) |
-
2022
- 2022-04-12 CN CN202210378351.0A patent/CN114937068A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9658443B2 (en) | Optics apparatus with detection of light rays received at different angles for output indicative of aliased views | |
CN105264361B (en) | High resolution computed tomography | |
CN107693037B (en) | PET scanning device and time offset correction method thereof | |
Goulet et al. | Novel, full 3D scintillation dosimetry using a static plenoptic camera | |
CN103607951A (en) | Image processor and image processing method | |
CN109166154A (en) | Light-field camera calibration method for light field three dimensional particles image reconstruction | |
CN103630091B (en) | Leaf area measurement method based on laser and image processing techniques | |
EP2702450B1 (en) | System and method for correction of vignetting effect in multi-camera flat panel x-ray detectors | |
CN112258593B (en) | CT or PET-CT intelligent positioning scanning method under monocular camera | |
US20130315454A1 (en) | Automated crystal identification achieved via watershed segmentation | |
Miranda et al. | Fast and accurate rat head motion tracking with point sources for awake brain PET | |
LaBella et al. | Convolutional neural network for crystal identification and gamma ray localization in PET | |
Rilling et al. | Tomographic‐based 3D scintillation dosimetry using a three‐view plenoptic imaging system | |
De Acilu et al. | Study and optimization of positioning algorithms for monolithic PET detectors blocks | |
Cloutier et al. | Deformable scintillation dosimeter: II. Real-time simultaneous measurements of dose and tracking of deformation vector fields | |
CN108209958A (en) | A kind of the determining of normalization factor, acquisition methods and medical imaging procedure | |
CN103228213B (en) | For monitoring the method for dose of radiation | |
CN112508171A (en) | Image depth estimation method and device based on multilayer convolutional neural network | |
Jung et al. | Compton background elimination for in vivo x-ray fluorescence imaging of gold nanoparticles using convolutional neural network | |
CN108983273B (en) | Real-time measurement system and method for three-dimensional dose distribution in human body | |
CN114937068A (en) | Method, system and device for obtaining tissue equivalent scintillator dose distribution | |
CN110796713B (en) | Radioactive source positioning method and device, electronic equipment and computer readable storage medium | |
Beattie et al. | Registration of planar bioluminescence to magnetic resonance and x-ray computed tomography images as a platform for the development of bioluminescence tomography reconstruction algorithms | |
US20230058112A1 (en) | Machine learning-based scintillator resonse modelling for increased spatial resolution in nuclear imaging | |
CN111839569B (en) | PET time resolution measuring, calculating and reconstructing system and method |
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 |