Phase contrast imaging method, device, equipment and storage medium

Abstract

The embodiment of the present invention discloses a phase contrast imaging method, device, equipment and storage medium, the method comprising: obtaining a single-step projection image corresponding to a single phase step projection data, obtaining a virtual step projection image of multiple virtual steps based on the single-step projection image; extracting material characteristic parameters of the virtual step projection image, obtaining a multi-step step synthesis projection image according to the material characteristic parameters of the virtual step projection image and the background phase information of the multi-step step; inputting the multi-step step synthesis projection image into a trained information extraction model to obtain optimized phase information output by the information extraction model; generating a target phase contrast image according to the material characteristic parameters and the optimized phase information. The embodiment of the present invention realizes that the phase information can be restored by only a single projection image containing the object by performing virtual stepping on the single-step projection image to extract phase information, and the accuracy and resolution of the optimized phase information are improved by optimizing the phase information based on the trained information extraction model.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of imaging technology, and in particular to a phase contrast imaging method, device, equipment and storage medium. Background Art

[0002] Phase contrast imaging is based on the principle that the phase change obtained by X-rays passing through light element materials is one thousand to one hundred thousand times the change in light intensity, which enables it to observe the internal structure of light element samples with much higher sensitivity than traditional absorption imaging. Therefore, it has a wide range of application scenarios in the fields of biomedicine and industrial detection. A major factor in X-ray grating phase contrast imaging is information recovery. At present, the information recovery of X-ray grating phase contrast imaging mainly includes phase stepping method, Fourier transform method and virtual stepping method. The phase stepping method is to obtain a multi-step object projection image and a corresponding background image by moving the grating (at least 3 steps) within a grating period, and then recover the absorption, phase and dark field information of the object; the Fourier transform method only needs to perform Fourier transform on the single-step object projection image and background image, and extract the absorption, phase and dark field information of the object from the first-order spectrum and the zero-order spectrum; the virtual stepping method is to translate a single projection image by multiple steps to simulate a multi-step image, and then obtain the absorption, phase and dark field information of the object. Among the above-mentioned information recovery methods, the phase stepping method requires the acquisition of multiple images, which will greatly increase the imaging time of the X-ray imaging system and increase the radiation dose received by the object. In addition, mechanical movement processes such as phase stepping also greatly reduce the stability of the system. The information recovered by the Fourier transform method and the virtual stepping method has serious artifacts and the spatial resolution is not high. Summary of the invention

[0003] The embodiments of the present invention provide a phase contrast imaging method, apparatus, device and storage medium, so as to restore phase information with good accuracy and resolution with only a single image.

[0004] In a first aspect, an embodiment of the present invention provides a phase contrast imaging method, comprising:

[0005] Acquire a single-step projection image corresponding to a single phase step projection data, and obtain a plurality of virtual step projection images of virtual steps based on the single-step projection image;

[0006] Extracting material characteristic parameters of the virtual step projection image, and obtaining a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and background phase information of the multi-step step;

[0007] Inputting the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model;

[0008] Generate a target phase contrast image based on material characteristic parameters and optimized phase information.

[0009] In a second aspect, an embodiment of the present invention further provides a phase contrast imaging device, comprising:

[0010] A virtual step projection image module is used to obtain a single-step projection image corresponding to a single phase step projection data, and obtain a plurality of virtual step projection images of virtual steps based on the single-step projection image;

[0011] A synthetic projection image module is used to extract material characteristic parameters of the virtual step projection image, and obtain a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and the background phase information of the multi-step step;

[0012] An optimized phase information module is used to input the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model;

[0013] The target phase contrast image module is used to generate a target phase contrast image according to material characteristic parameters and optimized phase information.

[0014] In a third aspect, an embodiment of the present invention further provides a computer device, the device comprising:

[0015] one or more processors;

[0016] A storage device for storing one or more programs;

[0017] When one or more programs are executed by one or more processors, the one or more processors implement the phase contrast imaging method provided by any embodiment of the present invention.

[0018] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a phase contrast imaging method as provided in any embodiment of the present invention.

[0019] The embodiment of the present invention obtains a single-step projection image corresponding to a single phase stepping projection data, and obtains a plurality of virtual stepping projection images based on the single-step projection image; extracts material characteristic parameters of the virtual stepping projection image, and obtains a multi-step stepping synthetic projection image according to the material characteristic parameters of the virtual stepping projection image and the background phase information of the multi-step stepping; inputs the multi-step stepping synthetic projection image into a trained information extraction model to obtain optimized phase information output by the information extraction model; generates a target phase contrast image according to the material characteristic parameters and the optimized phase information, and virtually steps the single-step projection image to extract the phase information, thereby realizing that the phase information can be restored by only a single projection image containing the object, and the phase information is optimized based on the trained information extraction model, thereby improving the accuracy and resolution of the extracted optimized phase information. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flow chart of a phase contrast imaging method provided by Embodiment 1 of the present invention;

[0021] Figure 2 is a flow chart of a phase contrast imaging method provided by Embodiment 2 of the present invention;

[0022] Figure 3a This is a schematic diagram of a training process of an information extraction model for phase contrast imaging provided by Embodiment 3 of the present invention;

[0023] Figure 3b is a schematic diagram of generating a sample multi-step synthetic projection image provided by the third embodiment of the present invention;

[0024] Figure 3c It is a structural diagram of an information extraction model provided by Embodiment 3 of the present invention;

[0025] Figure 3d is a training flow chart of the information extraction model provided in the third embodiment of the present invention;

[0026] Figure 4 is a structural schematic diagram of a phase contrast imaging device provided by Embodiment 4 of the present invention;

[0027] Figure 5 It is a structural diagram of a computer device provided in Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0028] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0029] Embodiment 1

[0030] Figure 1 1 is a flow chart of a phase contrast imaging method provided in the first embodiment of the present invention. This embodiment is applicable to the situation when phase contrast imaging is performed. The method can be executed by a phase contrast imaging device, and the phase contrast imaging device can be implemented in software and / or hardware. For example, the phase contrast imaging device can be configured in a computer device. Figure 1 As shown, the method includes:

[0031] S110, obtaining a single-step projection image corresponding to a single phase stepping projection data, and obtaining a plurality of virtual stepping projection images of virtual steps based on the single-step projection image.

[0032] In this embodiment, only the projection data of a single phase step need to be collected to obtain the target phase contrast image. Optionally, after obtaining the single-step projection image, the single-step projection image is translated in space to obtain multiple virtual step projection images of virtual steps. Among them, the translation of the single-step projection image in space can be: determine the number M of virtual step projection images, select M pixels based on the step mode with the center pixel of the single-step projection image as the starting point, and use the selected M pixels as the center pixels of each virtual step projection image to obtain M virtual step projection images.

[0033] S120, extracting material characteristic parameters of the virtual step projection image, and obtaining a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and background phase information of the multi-step step.

[0034] Optionally, after obtaining multiple virtual step projection images of virtual steps, the material characteristic parameters of the virtual step projection images are extracted, wherein the material characteristic parameters are various information of the object under the phase contrast imaging system, including object absorption information, object phase information and object scattering information, and then a multi-step synthetic projection image is obtained based on the object absorption information, object phase information and object scattering information and the background phase information of the multi-step steps. In the above process, the background phase information of the multi-step step can be obtained through the step background image of the actual multi-step step. Optionally, before obtaining the single phase step projection data of the object, the background image of the multi-step step is obtained, and the phase information in the multi-step step background image is extracted as the background phase information. Optionally, the method of extracting the background phase information can refer to the method of extracting the phase information in the prior art, such as the least squares method, the global transformation method, etc. Before the object is imaged for information acquisition, the background image of the multi-step step is obtained in advance to extract the background phase information without affecting the actual imaging process of the object, and the extracted background phase information can participate in the optimization of the object phase information.

[0035] When the material characteristic parameters are specifically object absorption information, object phase information and object scattering information, the multi-step synthetic projection image can be obtained according to the material characteristic parameters of the virtual step projection image and the background phase information of the multi-step stepping: Get a multi-step synthetic projection map. Among them, It is a multi-step synthetic projection map. For the object to absorb information, φ vps is the object phase information, ε vps is the object scattering information, φ air is the background phase information, and M is the number of virtual step projection images.

[0036] In one embodiment, extracting material characteristic parameters of the virtual step projection image includes: obtaining object absorption information, object phase information and object scattering information of the virtual step projection image by least square method. Optionally, there may be multiple ways to extract material characteristic parameters, such as least square method, global transformation method, etc. Preferably, the least square method is used to extract material characteristic parameters of the virtual step projection image, such as object absorption information, object phase information and object scattering information. The use of the least square method can make the extracted material characteristic parameters more accurate.

[0037] S130, inputting the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model.

[0038] In this embodiment, after obtaining the multi-step synthetic projection image, the multi-step synthetic projection image is input into the trained information extraction model, and the information extraction model is used to optimize the phase information extracted from the multi-step synthetic projection image to obtain the optimized phase information output by the information extraction model. The information extraction model is obtained by iteratively training the pre-constructed information extraction model with the sample multi-step synthetic projection image as the training sample and the mean square error between the predicted projection image and the sample projection image being less than the set error threshold as the goal.

[0039] S140, generating a target phase contrast image according to the material characteristic parameters and the optimized phase information.

[0040] After the optimized phase information is obtained, the target phase contrast image is obtained through the material characteristic parameters of the virtual step projection image and the optimized phase information. When the material characteristic parameters of the virtual step projection image are specifically object absorption information, object phase information and object scattering information, the target phase contrast image is generated according to the material characteristic parameters and the optimized phase information, including: generating the target phase contrast image according to the object absorption information, object scattering information and the optimized phase information. Specifically, the target phase contrast image is generated according to the object absorption information and object scattering information extracted from the virtual step projection image and the optimized phase information output by the information extraction model. It can be understood that the accuracy and resolution of the optimized phase information are good, so the target phase contrast image generated based on the optimized phase information is also more accurate and has a higher spatial resolution.

[0041] The embodiment of the present invention obtains a single-step projection image corresponding to a single phase stepping projection data, and obtains a plurality of virtual stepping projection images based on the single-step projection image; extracts material characteristic parameters of the virtual stepping projection image, and obtains a multi-step stepping synthetic projection image according to the material characteristic parameters of the virtual stepping projection image and the background phase information of the multi-step stepping; inputs the multi-step stepping synthetic projection image into a trained information extraction model to obtain optimized phase information output by the information extraction model; generates a target phase contrast image according to the material characteristic parameters and the optimized phase information, and virtually steps the single-step projection image to extract the phase information, thereby realizing that the phase information can be restored by only a single projection image containing the object, and the phase information is optimized based on the trained information extraction model, thereby improving the accuracy and resolution of the extracted optimized phase information.

[0042] Embodiment 2

[0043] Figure 2 1 is a flow chart of a phase contrast imaging method provided by the second embodiment of the present invention. This embodiment is further optimized on the basis of the above scheme. Figure 2 As shown, the method includes:

[0044] S210, obtaining a sample projection map, and obtaining a plurality of sample stepping projection maps of virtual stepping based on the sample projection map.

[0045] In this embodiment, a projection image corresponding to a single phase step projection data is obtained as a sample projection image, and a training sample of the information extraction model is generated based on the sample projection image. Specifically, the sample projection image is translated in space to obtain multiple virtual step sample step projection images. Among them, the method of translating the sample projection image in space can refer to the method of translating the single step projection image in space in the above embodiment, which will not be repeated here.

[0046] S220, extracting material characteristic parameters of the sample step projection image, and obtaining a sample multi-step synthetic projection image according to the material characteristic parameters of the sample step projection image and the multi-step sample background phase.

[0047] The material characteristic parameters of the sample step projection image may be the sample absorption information, sample phase information and sample scattering information of the sample step projection image. The material characteristic parameters of the sample step projection image may be extracted in a variety of ways, such as the least squares method, the global transformation method and the like. Optionally, the material characteristic parameters of the sample step projection image are extracted, including: obtaining the sample absorption information, sample phase information and sample scattering information of the sample step projection image by the least squares method. Preferably, the material characteristic parameters of the sample step projection image may be extracted by the least squares method. The use of the least squares method can make the extracted material characteristic parameters more accurate. Among them, the sample multi-step synthetic projection image is obtained according to the material characteristic parameters of the sample step projection image and the multi-step sample background phase, which can be: by Get the sample multi-step synthetic projection map. Among them, is the sample multi-step synthetic projection map, is the object absorption information in the material characteristic parameters of the sample step projection image, φ vps is the object phase information in the material characteristic parameters of the sample step projection image, ε vps is the object scattering information in the material characteristic parameters of the sample step projection image, φ air is the sample background phase of multiple steps.

[0048] Optionally, the sample background phase can be obtained by actually obtaining a multi-step stepped background image. Optionally, before obtaining the sample projection image, a multi-step stepped sample background image is obtained, and the phase information in the multi-step stepped sample background image is extracted as the sample background phase. Optionally, the method for extracting the sample background phase can refer to the method for extracting the background phase information in the above embodiment, which will not be repeated here.

[0049] S230, using the sample multi-step synthetic projection image as a training sample, with the goal of making the mean square error between the predicted projection image and the sample projection image less than a set error threshold, iteratively training the pre-constructed information extraction model to obtain a trained information extraction model.

[0050] In this embodiment, the information extraction model outputs predicted phase information, but in the training process of the information extraction model, the training is not performed with phase information as the target, but with sample projection images as the target. Specifically, the predicted phase information extracted by the information extraction model is combined with the material characteristic parameters of the sample without the projection image to obtain a predicted projection image, and the information extraction model is trained with the goal of the mean square error between the predicted projection image and the sample projection image being less than the set error threshold. It can be seen that the training samples of the information extraction model in this embodiment only include sample projection images and sample multi-step synthetic projection images, that is, in this embodiment, the training samples of the information extraction model can be obtained without labeling the image.

[0051] Optionally, the information extraction model may include a backbone network consisting of multiple 1*1 convolution kernels and a branch network consisting of multiple 1*n or n*1 convolution kernels, which is beneficial to improve the accuracy of the output information while ensuring the output image resolution.

[0052] In one embodiment, a sample multi-step synthetic projection image is used as a training sample, and a pre-constructed information extraction model is iteratively trained with the goal of making the mean square error between the predicted projection image and the sample projection image less than a set error threshold to obtain a trained information extraction model, including: inputting the sample multi-step synthetic projection image into the pre-constructed information extraction model to obtain predicted phase information output by the information extraction model; generating a predicted projection image based on the predicted phase information, sample absorption information, and sample scattering information; calculating the mean square error between the predicted projection image and the sample projection image, and iteratively training the pre-constructed information extraction model with the goal of making the mean square error less than a set error threshold to obtain a trained information extraction model. Specifically, the training process of the information extraction model can be: inputting the sample multi-step synthetic projection map into the pre-built information extraction model to obtain the predicted phase information output by the information extraction model; generating a predicted projection map according to the predicted phase information, sample absorption information and sample scattering information; calculating the mean square error between the predicted projection map and the sample projection map, and when the mean square error is less than the set error threshold, obtaining the trained information extraction model; when the mean square error is not less than the set error threshold, adjusting the parameter value in the information extraction model, re-extracting the predicted phase information, obtaining the predicted projection map, calculating the mean square error between the predicted projection map and the sample projection map, until the mean square error is less than the set error threshold or the number of iterations reaches the set number, and obtaining the trained information extraction model.

[0053] S240, obtaining a single-step projection image corresponding to a single phase stepping projection data, and obtaining a plurality of virtual stepping projection images of virtual steps based on the single-step projection image.

[0054] S250, extracting material characteristic parameters of the virtual step projection image, and obtaining a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and background phase information of the multi-step step.

[0055] S260, inputting the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model.

[0056] S270 , generating a target phase contrast image according to the material characteristic parameters and the optimized phase information.

[0057] The embodiment of the present invention obtains a trained information extraction model by training a pre-constructed information extraction model based on a sample projection map and a sample multi-step synthetic projection map, so that training samples of the information extraction model can be obtained without labeling the image, thereby solving the problem of being unable to obtain a large amount of labeled data required in the training process and improving the training efficiency of the information extraction model.

[0058] Embodiment 3

[0059] This embodiment provides a preferred embodiment based on the above embodiment. In this embodiment, the phase information extracted is optimized by training the information extraction model. Specifically, Figure 3a FIG. 1 is a schematic diagram of a training process of an information extraction model for phase contrast imaging provided by Embodiment 3 of the present invention. Figure 3a As shown, the method includes:

[0060] S310, obtaining a multi-step sample multi-step synthetic projection image based on a single sample projection image.

[0061] Figure 3b Schematic diagram of generating a sample multi-step synthetic projection image provided by the third embodiment of the present invention. Figure 3b As shown, firstly, a single sample projection image is By translating in space, we can obtain multiple virtual step sample step projection images. The corresponding sample absorption information can be restored by using the least squares method on the sample step projection images. Sample phase information φ vps and sample scattering information ε vps Then the sample background phase φ is extracted from the multi-step background image air The sample absorption information obtained above Sample phase information φ vps , sample scattering information ε vps and the sample background phase φ air Substituting into the physical imaging model, we can get the multi-step sample multi-step synthetic projection map Where k is a positive integer, M is the number of phase steps, and generally M≥3.

[0062] Among them, the physical imaging model can be

[0063] S320: training an information extraction model.

[0064] The sample multi-step synthetic projection image is used as the input of the information extraction model to train the information extraction model. Figure 3c Schematic diagram of the structure of an information extraction model provided by Embodiment 3 of the present invention. Figure 3cAs shown in the figure, the backbone of the information extraction model consists of a large number of 1*1 convolution kernels, and the branches are composed of 1*n and n*1 convolution kernels, which is conducive to improving the accuracy of the output information while ensuring the output image resolution. Specifically, the solid line blocks represent the convolution layer, and the numbers in the brackets represent the height of the convolution kernel, the width of the convolution kernel, the number of input channels, and the number of output channels, respectively. The activation function used is the leaky-relu activation function. The input of the information extraction model is the sample multi-step synthetic projection image The output is the predicted phase information φ net .

[0065] Figure 3d : is a training flow chart of the information extraction model provided in the third embodiment of the present invention. Figure 3d As shown, the sample multi-step synthesis projection map Input into the information extraction model to obtain the predicted phase information φ output by the information extraction model net , then based on the predicted phase information φ net , sample absorption information and sample scattering information ε vps Generate a predicted projection map of a single-step object By optimizing the predicted projection map With sample projection diagram I (k) The mean square error between makes the predicted phase of the information extraction model closer and closer to the actual phase information. Get the predicted projection map The information extraction model provided in this embodiment realizes the extraction of phase information from a single input projection image without the need for corresponding label data, solves the problem of being unable to obtain the large amount of data required for deep learning training, and avoids the tedious operation of obtaining a large amount of label data.

[0066] The embodiment of the present invention first uses a virtual stepping method to extract absorption, phase contrast and dark field information from a single projection image, extracts background phase information from a multi-step background image, and inputs the above four information into a physical imaging model to obtain a multi-step simulated projection image, and then inputs a simulated projection image into an information extraction model, and substitutes the phase information output by the information extraction model and the absorption, phase contrast and dark field information extracted from a single projection image into the physical imaging model to obtain a single predicted projection image. By reducing the mean square error between the predicted projection image and the original projection image, the phase information output by the network is made more accurate, and the phase information can be extracted using a single projection image. Compared with the phase stepping method, the method proposed in this embodiment can greatly shorten the imaging time, reduce the radiation dose received by the object, and has higher mechanical stability; compared with the Fourier transform method and the virtual stepping method, the method proposed in this embodiment can extract phase information with better accuracy and resolution.

[0067] Embodiment 4

[0068] Figure 4 1 is a schematic diagram of the structure of a phase contrast imaging device provided by the fourth embodiment of the present invention. The phase contrast imaging device can be implemented in software and / or hardware, for example, the phase contrast imaging device can be configured in a computer device. Figure 4 As shown, the device includes a virtual stepping projection image module 410, a synthetic projection image module 420, an optimized phase information module 430 and a target phase contrast image module 440, wherein:

[0069] A virtual step projection image module 410 is used to obtain a single-step projection image corresponding to a single phase step projection data, and obtain a plurality of virtual step projection images of virtual steps based on the single-step projection image;

[0070] The synthetic projection image module 420 is used to extract the material characteristic parameters of the virtual step projection image, and obtain the multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and the background phase information of the multi-step step;

[0071] The optimized phase information module 430 is used to input the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model;

[0072] The target phase contrast image module 440 is used to generate a target phase contrast image according to the material characteristic parameters and the optimized phase information.

[0073] The embodiment of the present invention obtains a single-step projection image corresponding to a single phase step projection data through a virtual step projection image module, and obtains a plurality of virtual step projection images of virtual steps based on the single-step projection image; the synthetic projection image module extracts the material characteristic parameters of the virtual step projection image, and obtains a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and the background phase information of the multi-step steps; the optimized phase information module inputs the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model; the target phase contrast image module generates a target phase contrast image according to the material characteristic parameters and the optimized phase information, and extracts the phase information by virtually stepping the single-step projection image, so that the phase information can be restored by only a single projection image containing the object, and the phase information is optimized based on the trained information extraction model, so as to improve the accuracy and resolution of the extracted optimized phase information.

[0074] Optionally, based on the above solution, the synthetic projection image module 420 is specifically used for:

[0075] The object absorption information, object phase information and object scattering information of the virtual step projection image are obtained by the least square method.

[0076] Optionally, based on the above solution, the target phase contrast image module 440 is specifically used for:

[0077] A target phase contrast image is generated according to the object absorption information, the object scattering information and the optimized phase information.

[0078] Optionally, based on the above solution, a model training module is also included, which is used to:

[0079] Acquire a sample projection map, and obtain multiple sample step projection maps of virtual stepping based on the sample projection map;

[0080] Extracting material characteristic parameters of the sample step projection image, and obtaining the sample multi-step synthetic projection image according to the material characteristic parameters of the sample step projection image and the multi-step sample background phase;

[0081] The sample multi-step synthetic projection image is used as a training sample. The pre-constructed information extraction model is trained with the goal of making the mean square error between the predicted projection image and the sample projection image less than the set error threshold to obtain a trained information extraction model, wherein the predicted projection image is obtained based on the extracted predicted phase information and the material characteristic parameters of the sample step projection image.

[0082] Optionally, based on the above solution, the model training module is specifically used for:

[0083] The sample absorption information, sample phase information and sample scattering information of the sample step projection image are obtained by the least squares method.

[0084] Optionally, based on the above solution, the model training module is specifically used for:

[0085] Inputting the sample multi-step synthetic projection image into a pre-built information extraction model to obtain the predicted phase information output by the information extraction model;

[0086] Generate a predicted projection map based on predicted phase information, sample absorption information, and sample scattering information;

[0087] The mean square error between the predicted projection map and the sample projection map is calculated. With the goal of making the mean square error less than the set error threshold, the pre-built information extraction model is trained to obtain a trained information extraction model.

[0088] Optionally, based on the above scheme, the information extraction model includes a trunk network composed of multiple 1*1 convolution kernels and a branch network composed of multiple 1*n or n*1 convolution kernels.

[0089] The phase contrast imaging device provided in the embodiment of the present invention can execute the phase contrast imaging method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0090] Embodiment 5

[0091] Figure 5 It is a structural diagram of a computer device provided in Embodiment 5 of the present invention. Figure 5 A block diagram of an exemplary computer device 512 suitable for use in implementing embodiments of the present invention is shown. Figure 5 The computer device 512 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0092] like Figure 5 As shown, the computer device 512 is in the form of a general-purpose computing device. The components of the computer device 512 may include, but are not limited to: one or more processors 516, a system memory 528, and a bus 518 that connects various system components (including the system memory 528 and the processor 516).

[0093] Bus 518 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a local bus to processor 516, or a bus structure using any of a variety of bus architectures. These architectures include, by way of example, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0094] The computer device 512 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 512, including volatile and non-volatile media, removable and non-removable media.

[0095] System memory 528 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 530 and / or cache memory 532. Computer device 512 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage device 534 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 518 via one or more data medium interfaces. The memory 528 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0096] A program / utility 540 having a set (at least one) of program modules 542 may be stored, for example, in the memory 528, such program modules 542 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 542 generally perform the functions and / or methods of the embodiments described herein.

[0097] The computer device 512 may also communicate with one or more external devices 514 (e.g., keyboard, pointing device, display 524, etc.), may also communicate with one or more devices that enable a user to interact with the computer device 512, and / or communicate with any device that enables the computer device 512 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed via an input / output (I / O) interface 522. In addition, the computer device 512 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) via a network adapter 520. As shown, the network adapter 520 communicates with other modules of the computer device 512 via a bus 518. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the computer device 512, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0098] The processor 516 executes various functional applications and data processing by running the program stored in the system memory 528, for example, implementing the phase contrast imaging method provided in the embodiment of the present invention, which includes:

[0099] Acquire a single-step projection image corresponding to a single phase step projection data, and obtain a plurality of virtual step projection images of virtual steps based on the single-step projection image;

[0100] Extracting material characteristic parameters of the virtual step projection image, and obtaining a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and background phase information of the multi-step step;

[0101] Inputting the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model;

[0102] Generate a target phase contrast image based on material characteristic parameters and optimized phase information.

[0103] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the phase contrast imaging method provided by any embodiment of the present invention.

[0104] Embodiment 6

[0105] Embodiment 6 of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the phase contrast imaging method provided in the embodiment of the present invention is implemented. The method includes:

[0106] Acquire a single-step projection image corresponding to a single phase step projection data, and obtain a plurality of virtual step projection images of virtual steps based on the single-step projection image;

[0107] Extracting material characteristic parameters of the virtual step projection image, and obtaining a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and background phase information of the multi-step step;

[0108] Inputting the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model;

[0109] Generate a target phase contrast image based on material characteristic parameters and optimized phase information.

[0110] Of course, the computer program stored in a computer-readable storage medium provided by an embodiment of the present invention is not limited to the above method operations, but can also execute related operations of the phase contrast imaging method provided by any embodiment of the present invention.

[0111] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. Computer-readable media can be computer-readable information media or computer-readable storage media. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: electrical connections with one or more wires, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this document, computer-readable storage media can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, device or device.

[0112] Computer-readable information media may include data information propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data information may take a variety of forms, including but not limited to electromagnetic information, optical information, or any suitable combination of the above. Computer-readable information media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0113] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0114] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0115] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A phase contrast imaging method, characterized in that: include: Acquire a single-step projection image corresponding to a single phase step projection data, and obtain a plurality of virtual step projection images of virtual steps based on the single-step projection image; Extracting material characteristic parameters of the virtual step projection image, and obtaining a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and background phase information of the multi-step step; Inputting the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model; A target phase contrast image is generated according to the material characteristic parameters and the optimized phase information.

2. The method according to claim 1, characterized in that: The step of extracting material characteristic parameters of the virtual stepping projection image comprises: The object absorption information, object phase information and object scattering information of the virtual step projection image are obtained by the least square method.

3. The method according to claim 2, characterized in that The generating a target phase contrast image according to the material characteristic parameters and the optimized phase information comprises: The target phase contrast image is generated according to the object absorption information, the object scattering information and the optimized phase information.

4. The method according to claim 3, characterized in that The information extraction model includes a trunk network composed of multiple 1*1 convolution kernels and a branch network composed of multiple 1*n or n*1 convolution kernels.

5. The method according to claim 1, characterized in that Also includes: Acquire a sample projection image, and obtain multiple sample step projection images of virtual steps based on the sample projection image; Extracting material characteristic parameters of the sample step projection image, and obtaining a sample multi-step synthetic projection image according to the material characteristic parameters of the sample step projection image and a multi-step sample background phase; The sample multi-step synthetic projection image is used as a training sample, and the pre-constructed information extraction model is trained with the goal of making the mean square error between the predicted projection image and the sample projection image less than a set error threshold to obtain a trained information extraction model, wherein the predicted projection image is obtained based on the extracted predicted phase information and the material characteristic parameters of the sample step projection image.

6. The method according to claim 5, characterized in that The step of extracting material characteristic parameters of the sample step projection image includes: The sample absorption information, sample phase information and sample scattering information of the sample step projection image are obtained by the least square method.

7. The method according to claim 5, characterized in that The method uses the sample multi-step synthetic projection image as a training sample, trains the pre-built information extraction model with the goal of making the mean square error between the predicted projection image and the sample projection image less than a set error threshold, and obtains the trained information extraction model, including: Inputting the sample multi-step synthetic projection image into a pre-built information extraction model to obtain predicted phase information output by the information extraction model; generating a predicted projection map according to the predicted phase information, the sample absorption information and the sample scattering information; The mean square error between the predicted projection image and the sample projection image is calculated, and the pre-constructed information extraction model is trained with the goal of making the mean square error smaller than a set error threshold to obtain a trained information extraction model.

8. A phase contrast imaging device, characterized in that: include: A virtual step projection image module, used to obtain a single-step projection image corresponding to a single phase step projection data, and obtain a plurality of virtual step projection images of virtual steps based on the single-step projection image; A synthetic projection image module, used to extract material characteristic parameters of the virtual step projection image, and obtain a multi-step synthetic projection image according to the material characteristic parameters of the virtual step projection image and the background phase information of the multi-step steps; An optimized phase information module, used for inputting the multi-step synthetic projection image into the trained information extraction model to obtain the optimized phase information output by the information extraction model; The target phase contrast image module is used to generate a target phase contrast image according to the material characteristic parameters and the optimized phase information.

9. A computer device, characterized in that: The device comprises: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the phase contrast imaging method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the phase contrast imaging method as described in any one of claims 1 to 7 is implemented.

Images