Image reconstruction method, device, electronic device and storage medium

Abstract

The present disclosure relates to an image reconstruction method, an image reconstruction device, an electronic device, a computer-readable storage medium, and a computer program product. The image reconstruction method includes: acquiring image data of a computer tomography scan; based on the image data, using a trained neural network model to calculate an output image, wherein the trained neural network model is configured to be trained via an image database including at least one training image pair consisting of a first image and a second image, and wherein the first image and the second image in the training image pair are respectively used as a training input image and a training output image, and the first image and the second image are configured to be obtained at the same time point via computer tomography of the same biological tissue. According to the image reconstruction method of the present disclosure, a more accurate, reliable, and high-quality image can be calculated.

Description

Technical Field

[0001] The present disclosure relates to the field of medical device technology, and in particular to an image reconstruction method, device, electronic device, computer-readable storage medium, and computer program product. Background Art

[0002] Medical image imaging and reconstruction technology is widely used in medical detection. Among them, image imaging and reconstruction based on computed tomography (CT) detects the local X-ray attenuation of the patient's body structure by recording time integral. It has the characteristics of fast scanning time and clear image, and has attracted widespread attention.

[0003] The increase in the use of CT has led to widespread concern about the high radiation dose of CT examinations relative to other X-ray examinations. Several studies have focused on how to reduce patient dose and operate under the ALARA principle.

[0004] The methods described in this section are not necessarily methods that have been previously conceived or employed. Unless otherwise indicated, it should not be assumed that any method described in this section is considered to be prior art simply because it is included in this section. Similarly, unless otherwise indicated, the issues mentioned in this section should not be considered to have been recognized in any prior art. Summary of the invention

[0005] According to one aspect of an embodiment of the present disclosure, an image reconstruction method is proposed, comprising: acquiring image data of a computed tomography scan; and calculating an output image based on the image data using a trained neural network model, wherein the trained neural network model is configured to be trained via an image database comprising at least one training image pair consisting of a first image and a second image, and wherein the first image and the second image in the training image pair serve as a training input image and a training output image, respectively, and the first image and the second image are configured to be obtained via a computed tomography scan of the same biological tissue at the same point in time.

[0006] According to another aspect of an embodiment of the present disclosure, an image reconstruction device is proposed, comprising: an acquisition unit configured to acquire image data of a computed tomography scan; and a processing unit configured to process the image data using a trained neural network model to obtain an output image based on the image data, wherein the trained neural network model is configured to be trained via an image database comprising at least one training image pair consisting of a first image and a second image, and wherein the first image and the second image in the training image pair serve as a training input image and a training output image, respectively, and the first image and the second image are configured to be images of the same biological tissue obtained via computed tomography at the same point in time.

[0007] According to another aspect of an embodiment of the present disclosure, an electronic device is proposed, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program, and the computer program implements a method according to an embodiment of the present disclosure when executed by the at least one processor.

[0008] According to another aspect of an embodiment of the present disclosure, a non-transitory computer-readable storage medium storing a computer program is provided, wherein the computer program implements a method according to an embodiment of the present disclosure when executed by a processor.

[0009] According to another aspect of an embodiment of the present disclosure, a computer program product is provided, including a computer program, wherein the computer program implements the method according to an embodiment of the present disclosure when executed by a processor.

[0010] According to one or more embodiments of the present disclosure, by using image pairs obtained by performing computer tomography on the same biological tissue at the same time point as training images for training a neural network, the positions of the scanned images are made consistent while avoiding the introduction of unnecessary noise. The neural network trained based on the training images can calculate more accurate and reliable high-quality images.

[0011] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings exemplarily illustrate the embodiments and constitute a part of the specification, and together with the text description of the specification, are used to explain the exemplary implementation of the embodiments. The embodiments shown are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0013] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings so that the above and other features and advantages of the present disclosure will be more apparent to those skilled in the art. In the accompanying drawings:

[0014] Figure 1 is an exemplary flow chart of an image reconstruction method according to an embodiment of the present disclosure;

[0015] Figure 2 is an exemplary flow chart of a method for acquiring computer tomography image data according to an embodiment of the present disclosure;

[0016] Figure 3 A schematic diagram of acquiring a training image pair via dual-source computed tomography according to an embodiment of the present disclosure;

[0017] Figure 4 is a schematic block diagram of an image reconstruction device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0018] In order to have a clearer understanding of the technical features, purposes and effects of the present disclosure, specific embodiments of the present disclosure are now described with reference to the accompanying drawings, in which the same reference numerals represent the same parts.

[0019] In this document, “exemplary” means “serving as an example, instance or illustration”, and any diagram or implementation described in this document as “exemplary” should not be interpreted as a more preferred or more advantageous technical solution.

[0020] In order to simplify the drawings, only the parts related to the present disclosure are schematically shown in each figure, and they do not represent the actual structure of the product. In addition, in order to simplify the drawings and facilitate understanding, in some figures, only one of the parts with the same structure or function is schematically drawn or marked.

[0021] In this article, "one" not only means "only one", but also means "more than one". In this article, "first", "second", etc. are only used to distinguish each other, rather than to indicate their importance and order, or the premise of each other's existence.

[0022] Computed tomography (CT) is a process of scanning a part of the human body section by section by section with a precisely collimated X-ray beam. The computed tomography (CT) system that performs computed tomography is based on volumetric data acquisition, where an X-ray tube and a receiver rotate around the patient to collect transmission data from a volume of tissue. The CT system mainly includes: a data acquisition unit, an image reconstruction unit, an image display unit, and a storage and communication unit. The data acquisition unit includes the use of an X-ray tube coupled to a special electronic receiver to systematically collect attenuation data in the form of a patient projection profile. The image reconstruction unit is based on a terminal computer and is used to execute complex computer algorithms to create images using projection profiles for images acquired via the data acquisition unit. The storage and communication unit, also known as "Picture Archiving and Communication Systems (PACS)", stores the images created by the image reconstruction unit and communicates them to remote locations.

[0023] The image reconstruction unit reconstructs the image using algorithms including a filtered back projection (FBP) algorithm and an iterative reconstruction algorithm.

[0024] The process of reconstructing an image based on the filtered back projection algorithm includes: applying a convolution filter to each set of projection data before back projection, and then back-projecting the filtered projections to generate a reconstructed image. The filtered back projection algorithm obtains reconstructed images, which have the advantages of short reconstruction time and image texture that adapts to the needs of radiologists in the image interpretation process. However, since the convolution filter introduces noise and artifacts into the image, it is difficult to obtain images at low doses to meet image quality requirements.

[0025] The process of reconstructing an image based on an iterative reconstruction algorithm is a computational process that uses a series of repeated operations. The iterative reconstruction algorithm constructs an image based on the acquired computed tomography data with a better defined mathematical estimate (system model) and multiple iterations. Among them, each iteration process undergoes a noise reduction. The iterative reconstruction algorithm maintains spatial resolution and image contrast while reducing image noise. Especially in the case of low tube current in low-dose CT examinations, by introducing an iterative reconstruction algorithm, the problem of easy introduction of noise and artifacts in the image reconstruction based on the FBP algorithm in low-dose CT scanning can be solved, and the detectability of spatial resolution and low contrast can be maintained even at low doses. However, the iterative reconstructed images obtained by the iterative reconstruction algorithm also have some problems, such as long reconstruction time and the appearance of spots, plastic appearance or unnatural noise textures, which have adverse effects on radiologists in the image interpretation process. Moreover, compared with the CT reconstructed images obtained based on the filtered back projection algorithm, the reconstructed image texture obtained based on the iterative reconstruction algorithm cannot adapt to the diagnosis and treatment habits and clinical experience of radiologists, which has a serious impact on the accuracy of the doctor's diagnosis process.

[0026] Based on this, an image reconstruction method based on machine learning is proposed. In the image reconstruction method based on machine learning, a neural network is usually used to learn the reconstructed image. The training of the neural network requires a large amount of data. The training method uses a low-dose CT reconstructed image as input and a high-dose CT reconstructed image of the same patient as output, so as to train a trained neural network suitable for calculating and outputting a high-dose CT reconstructed image through a low-dose CT reconstructed image. In order to obtain a low-dose CT reconstructed image and a high-dose CT reconstructed image, scanning can be performed at a high dose to obtain high-dose data, and low-dose simulated data can be obtained by artificially adding noise to the high-dose original data (virtual data); or low-dose and high-dose CT reconstructed images are collected by repeated scanning twice to obtain training images. However, the method of obtaining a low-dose simulated image by artificially adding noise often introduces redundant uncontrollable noise, resulting in different noises in the trained neural network; and the method of collecting low-dose and high-dose images by scanning twice may not only introduce redundant noise, but also may cause image position mismatch or artifacts.

[0027] According to one aspect of an embodiment of the present disclosure, a method for reconstructing an image based on a neural network using machine learning is provided. The embodiments of the present disclosure will be described in detail below in conjunction with the accompanying drawings.

[0028] See also Figure 1 The image reconstruction method 100 according to the embodiment of the present disclosure is exemplarily described. Figure 1 As shown, the image reconstruction method 100 includes:

[0029] Step 110: Acquire computer tomography image data;

[0030] Step 120: Based on the image data, an output image is calculated using a trained neural network model, wherein the trained neural network model is configured to be trained via an image database including at least one training image pair consisting of a first image and a second image, and wherein the first image and the second image in the training image pair serve as a training input image and a training output image, respectively, and the first image and the second image are configured to be obtained via computed tomography scanning of the same biological tissue at the same time point.

[0031] Neural network models include, for example, convolutional neural networks (CNN). Convolutional neural networks first perform a series of convolution steps and pooling steps, and then deconvolute the data obtained therefrom again and perform a "de-pooling" step. Convolution and deconvolution are performed with the aid of matrices, which are also referred to as "kernels". There can be an approximately one-dimensional range between convolution and deconvolution, which is usually referred to as a fully connected layer. The output image is calculated using a convolutional neural network using a second image as a training image, for example, with the aid of the values ​​obtained after the last pooling, a probability value for a classified object can be assigned, and the deconvolution process is assigned to the classified object reorganization in a manner corresponding to the first image to be input to calculate the output image.

[0032] In step 110, image data is acquired using a computed tomography (CT) system. In some embodiments, the image data acquired in step 110 includes a reconstructed image.

[0033] See below Figure 2 According to an exemplary embodiment of the present disclosure, the step 110 of acquiring the image data of the computer tomography is exemplarily introduced. According to some embodiments, acquiring the image data of the computer tomography includes:

[0034] Step 210: Acquire computer tomography data;

[0035] Step 220: Perform calculation based on the acquired computed tomography data to obtain the reconstructed image.

[0036] Exemplarily, the data of the computer tomography includes data converted from X-rays collected by the receiver during the computer tomography process. Specifically, during the computer tomography process, the X-rays transmitted by the computer tomography system after tomography of the biological tissue are converted into electrical signals through photoelectric conversion, and the electrical signals are then converted into digital signals through an analog / digital converter, and the digital signals are input into a computer for processing. Among them, the data of the computer tomography include the electrical signals, digital signals, etc. converted from the X-rays collected by the receiver.

[0037] It should be noted that in step 110, the image data obtained by the computer tomography system can be any type of computer scanning system, such as a single-source, dual-source, or multi-source computer tomography system, and is not limited here.

[0038] In some examples, the reconstructed image includes a filtered back projection reconstructed image obtained based on a filtered back projection (FBP) algorithm. In these examples, the filtered back projection reconstructed image is used as the image data of the computed tomography, which can be a reconstructed image based on a filtered back projection algorithm of a low-dose computed tomography obtained under low-dose conditions.

[0039] By using the method according to the embodiment of the present disclosure, a trained neural network model is used to calculate an output image using a low-dose filtered back projection reconstruction image as an input image, and the output image simulates a filtered back projection reconstruction image obtained in a high-dose computed tomography scan, thereby achieving a high-quality image with image clarity comparable to the image quality of a high-dose computed tomography scan image under low-dose scanning conditions. On the one hand, a lower-dose computed tomography scan process can be used for patients in a computed tomography scan examination to reduce the radiation dose to the patient, and on the other hand, it helps doctors obtain clearer computed tomography scan reconstruction images, assisting doctors in making more accurate analysis and judgments.

[0040] In some other embodiments, the reconstructed image includes an iteratively reconstructed image obtained based on an iterative reconstruction algorithm. In these examples, the iteratively reconstructed image is used as image data of a computed tomography scan, which can be an iteratively reconstructed image based on an iterative reconstruction algorithm of a low-dose computed tomography scan obtained under low-dose conditions.

[0041] Through the method of the disclosed embodiment, a trained neural network model is used to iteratively reconstruct images at low doses to calculate output images, and the output images simulate filtered back-projection reconstructed images obtained in high-dose computed tomography, so that high-quality images with image texture comparable to the image quality of high-dose computed tomography images can be obtained under low-dose scanning conditions. On the one hand, a lower-dose computed tomography scanning process is used for patients in computed tomography examinations to reduce the radiation dose to patients, and on the other hand, it helps doctors obtain reconstructed images that are more suitable for their recognition, identification and judgment in terms of texture, and assists doctors in making more accurate analysis and judgments.

[0042] In step 120, an output image is calculated based on the image data using a trained neural network model, wherein the trained neural network model is configured to be trained via an image database including at least one training image pair consisting of a first image and a second image, and wherein the first image and the second image in the training image pair are respectively used as a training input image and a training output image, and the first image and the second image are configured to be obtained via computer tomography of the same biological tissue at the same time point.

[0043] It should be understood that in the description of the present disclosure, the biological tissue may be an organ, a body part, or a region of interest of a patient to be scanned by a computed tomography scan, and is not limited thereto.

[0044] In a method according to an exemplary embodiment of the present disclosure, an image database including at least one training image pair consisting of a first image and a second image is used to train a neural network, wherein the first image and the second image are respectively used as a training input image and a training output image, and the first image and the second image are both images obtained by computer tomography, and the first image and the second image are obtained by computer tomography of the same biological tissue at the same time point, so that the first image and the second image obtained by computer tomography have the same noise, which can avoid introducing new noise compared to obtaining training images by adding noise. At the same time, the first image and the second image are obtained by computer tomography of the same biological tissue at the same time, and the position of the biological tissue in the first image and the second image is consistent, which avoids the introduction of redundant noise and the inconsistent position of the biological tissue in the image when the scanned image is obtained by scanning the biological tissue twice, resulting in the introduction of new variables in the process of training the neural network.

[0045] Since no new noise or new variables are introduced in the process of training the neural network by using the first image and the second image obtained by computer tomography of the same biological tissue at the same time point as training images, the trained neural network has high reliability and can calculate more accurate output results when the computer tomography image data is used as input data to calculate output data through the trained neural network.

[0046] In some embodiments, the training image pair includes a first image and a second image acquired at different doses, wherein the dose when the first image is acquired is lower than the dose when the second image is acquired.

[0047] A first image and a second image obtained by performing a computer tomography scan on the same biological tissue at the same time point are used as input images and output images to train a neural network, respectively, wherein the first image is an image obtained by computer tomography under low-dose conditions, and the second image is an image obtained by computer tomography under high-dose conditions, so that the trained neural network can calculate a clearer image with less noise by using the image obtained by computer tomography under low-dose conditions as an input image. The clearer image with less noise is equivalent to the image obtained by computer tomography under high-dose conditions, so that the radiation dose to the patient can be reduced in the computer tomography scan, and a low-dose computer tomography image can be obtained. Through the calculation of the neural network, an image that can assist doctors in making judgments is obtained.

[0048] It should be understood that in the embodiments of the present disclosure, the first image and the second image obtained by performing a computed tomography scan on the same biological tissue at the same time point are used as the input image and the output image to train the neural network, respectively. The trained neural network can be applied to any computed tomography system, for example, a single-source, dual-source, or multi-source computed tomography system.

[0049] In some embodiments, the training image pair includes a first image and a second image obtained at the same energy. The energy of the computed tomography scan refers to the tube voltage of the tube in the computed tomography scan system, which determines the CT value of the scanned object. By analyzing the CT value, a doctor can diagnose whether the patient's biological tissue has a lesion.

[0050] Since X-ray beams have radiation effects on the human body, it is often necessary to find a balance between radiation dose and accurate analysis of CT values ​​in computed tomography. Radiation dose is a parameter determined by the tube voltage and tube current of the tube in the computed tomography system. When either the tube current or the tube voltage increases, the radiation dose increases. The tube current has a significant effect on the noise of the computed tomography. When the tube current is low, the noise is high, which affects the analysis process of the CT value in the final computed tomography reconstructed image.

[0051] In the above embodiment, a first image and a second image are obtained at different doses under the same energy, for example, at different tube currents (mAs) under the same tube voltage (kV) of the X-ray tube of a computed tomography system. The first image and the second image are images with different noise levels under the same noise background.

[0052] The neural network trained by the first image and the second image can perform noise reduction processing on the image obtained at a low dose, thereby outputting an output image with a low noise level equivalent to that obtained at a high dose, which can be used as a basis for the doctor's diagnosis. Thus, the radiation dose to the patient in the computed tomography scan can be reduced, while being able to obtain sufficiently clear, accurate and low-noise images.

[0053] It should be understood that the same noise background means that the noise background from the biological tissue itself is the same, which is related to the biological tissue and the state of the biological tissue itself during computer scanning. Different biological tissues have different noise backgrounds during computer scanning, and the same biological tissue has different noise backgrounds during computer tomography scanning at different times. Having different noise levels under the same noise background means that the image has noise of different intensities due to different settings of tube current parameters during computer tomography scanning.

[0054] In an embodiment of the present disclosure, a computed tomography scan is performed on the same biological tissue at the same time point under different doses and with the same tube voltage and different tube current settings for the computed tomography parameters to obtain a first image and a second image. Since the first image and the second image are obtained by performing a computed tomography scan on the same biological tissue at the same time point, they have the same noise background; and since the first image and the second image are obtained by performing a computed tomography scan under different tube current parameters, they have noises of different intensities.

[0055] Since the first image and the second image are obtained by computer tomography scanning of the same biological tissue at the same time point, they have the same noise background and no additional noise is introduced. In the process of training the neural network, only the noise caused by the scanning parameter setting during the scanning process is processed by the neural network, thereby improving the accuracy of the neural network training. In some embodiments, the training image pair includes the first image and the second image obtained by a multi-source computer tomography (CT) system. For example, the training image is the first image and the second image obtained by dual-source computer tomography. See Figure 3 The first image and the second image obtained by dual-source computed tomography according to an embodiment of the present disclosure are exemplarily introduced. Figure 3 , which shows a schematic diagram of a dual-source computed tomography system according to an embodiment of the present disclosure.

[0056] The dual-source computed tomography system includes a first probe 301 and a second probe 302 for simultaneously generating an X-ray beam for tomography, and a first receiver 303 opposite to the first probe 301 for receiving the X-ray beam emitted by the first probe 301 and a first receiver 304 opposite to the second probe 302 for receiving the X-ray beam emitted by the second probe 302.

[0057] During the dual-source computed tomography scanning process, the X-ray beam emitted by the first probe 301 is used to scan the biological tissue to be measured, and the first receiver 303 receives the X-ray beam emitted by the first probe 301 that passes through the biological tissue. At the same time, the second probe 302 emits an X-ray beam to scan the biological tissue to be measured, and the second receiver 304 receives the X-ray beam emitted by the second probe 302 that passes through the biological tissue. Since the first probe 301 and the second probe 302 both emit X-ray beams to scan the same biological tissue at the same time, the difference in the X-ray beams that pass through the biological tissue and are received by the respective opposite receivers after scanning is only related to the parameters of the X-ray beams emitted by the first probe 301 and the second probe 302. Therefore, the noise caused by the biological tissue in the energy of the X-ray beams received by the first receiver 303 and the second receiver 304 is the same, and the noise difference of the reconstructed image reconstructed based on the energy of the X-ray beams received by both is only due to the difference in the parameters of the X-rays emitted by the first probe 301 and the second probe 302. Therefore, the reconstructed image reconstructed based on the energy of the X-rays emitted by one of the probes after being received can be subjected to noise reduction processing according to the reconstructed image reconstructed based on the energy of the X-rays emitted by the other probe after being received.

[0058] Continue to view Figure 3 , an exemplary introduction is given to obtaining a first image and a second image at different doses via dual-source computed tomography.

[0059] like Figure 3 As shown, the parameters of the X-ray beam emitted by the first probe 301 and the second probe 302 are set under different dose parameters. In one example, the tube voltage in the first probe 301 and the tube voltage in the second probe 302 are set to be the same, and the tube current in the second probe 302 is set to be larger than the tube current in the first probe 301, so that the second probe 302 emits X-rays to scan the biological tissue under test at a higher dose, while the first probe 301 emits X-rays to scan the biological tissue under test at a lower dose. Thus, a first image 305 reconstructed based on the energy received by the first receiver 303 and a second image 306 reconstructed based on the energy received by the second receiver 304 are obtained. Wherein, as Figure 3As shown, the first image 305 is noisier and less clear than the second image 306 .

[0060] Since the first probe 301 and the second probe 302 both emit X-ray beams to the same biological tissue at the same time to respectively scan the first image 305 and the second image 306, the noise difference in the first image 305 and the second image 306 is only related to the dose parameter settings in the first probe 301 and the second probe 302. Therefore, the neural network trained based on the first image 305 and the second image 306 can perform noise reduction processing on the low-dose reconstructed image without introducing noise that is unrelated to the dose parameters (such as the noise of the biological tissue, etc.).

[0061] It should be understood that, in the above-mentioned embodiment, the first image and the second image at different doses are obtained by dual-source computed tomography, which is only exemplary, and can also be obtained by multi-source computed tomography, which is not limited here. At the same time, it should be understood that, in the above-mentioned embodiment, the first image and the second image at different doses are obtained by dual-source computed tomography, which are only used as training images to train the neural network, and in the specific application process of the trained neural network, its input image can be image data obtained by any computed tomography system, for example, a single-source, dual-source, multi-source-computed tomography system, etc. For example, using the trained neural network, the image obtained by a single-source-computed tomography system at a low dose can be subjected to noise reduction processing to obtain a clearer image with less noise, and the clearer image with less noise is equivalent to the image obtained by a single-source-computed tomography system at a high dose.

[0062] In some embodiments, the first image in the image data pair used for training is configured as a filtered back-projection reconstructed image, and the second image is configured as a filtered back-projection reconstructed image.

[0063] In other embodiments, the first image in the image data pair used for training is configured as an iteratively reconstructed image, and the second image is configured as a filtered back-projection reconstructed image. The iteratively reconstructed image is used as the input image for training the neural network, and the filtered back-projection reconstructed image is used as the output image for training the neural network, so that the trained neural network can calculate an output image having the same texture as the filtered back-projection reconstructed image based on the iteratively reconstructed image, and the output image has the texture of the filtered back-projection reconstructed image, which conforms to the doctor's analysis and judgment habits, thereby helping the doctor to make more accurate analysis and judgment based on the reconstructed image that is more suitable for his recognition, identification and judgment in terms of texture.

[0064] According to another aspect of the present disclosure, an image reconstruction device is provided. Figure 4As shown, the image reconstruction device 400 may include: an acquisition unit 410, configured to acquire image data of a computed tomography scan; a processing unit 420, configured to process the image data using a trained neural network model to obtain an output image, wherein the trained neural network model is configured to be trained via an image database including at least one training image pair consisting of a first image and a second image, and wherein the first image and the second image in the training image pair serve as a training input image and a training output image, respectively, and the first image and the second image are configured to be images of the same biological tissue obtained via a computed tomography scan at the same time point.

[0065] In some embodiments, the training image pair includes a first image and a second image acquired at different doses, wherein the dose when the first image is acquired is lower than the dose when the second image is acquired. In some examples, the first image and the second image are acquired via, for example, Figure 3 The images shown were acquired using a dual-source computed tomography system.

[0066] In some embodiments, the image reconstruction apparatus 400 may be a single-source computed tomography system or a dual-source computed tomography system.

[0067] According to another aspect of an embodiment of the present disclosure, there is provided an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program, and the computer program implements the above-mentioned image reconstruction method when executed by the at least one processor.

[0068] In some embodiments, the electronic device may include a computed tomography system.

[0069] According to another aspect of an embodiment of the present disclosure, a non-transitory computer-readable storage medium storing a computer program is provided, wherein the computer program implements the above method when executed by a processor.

[0070] According to another aspect of an embodiment of the present disclosure, a computer program product is provided, including a computer program, wherein the computer program implements the above method when executed by a processor.

[0071] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0072] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0073] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0074] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0075] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0076] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0077] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0078] Although the embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it should be understood that the above-mentioned methods, systems and devices are merely exemplary embodiments or examples, and the scope of the present invention is not limited by these embodiments or examples, but only by the claims after authorization and their equivalent scope. Various elements in the embodiments or examples can be omitted or replaced by their equivalent elements. In addition, each step can be performed in an order different from that described in the present disclosure. Further, the various elements in the embodiments or examples can be combined in various ways. It is important that with the evolution of technology, many elements described herein can be replaced by equivalent elements that appear after the present disclosure.

Claims

1. An image reconstruction method, comprising: acquiring computed tomography image data; Based on the image data, an output image is calculated using a trained neural network model, wherein the trained neural network model is configured to be trained via an image database including at least one training image pair consisting of a first image and a second image, and wherein the first image and the second image in the training image pair serve as a training input image and a training output image, respectively, and the first image and the second image are configured to be obtained via computer tomography of the same biological tissue at the same time point; wherein the training image pair comprises the first image and the second image obtained via dual-source computed tomography; The dual-source computer tomography system includes a first probe and a second probe for simultaneously generating an X-ray beam for tomography, and a first receiver opposite to the first probe for receiving the X-ray beam emitted by the first probe and a first receiver opposite to the second probe for receiving the X-ray beam emitted by the second probe; During the dual-source computed tomography scanning process, the X-ray beam emitted by the first probe scans the biological tissue to be measured, and the first receiver receives the X-ray beam emitted by the first probe that passes through the biological tissue. At the same time, the second probe emits an X-ray beam to scan the biological tissue to be measured, and the second receiver receives the X-ray beam emitted by the second probe that passes through the biological tissue.

2. The image reconstruction method according to claim 1, wherein: The training image pair includes the first image and the second image acquired at different doses, wherein the dose when the first image is acquired is lower than the dose when the second image is acquired.

3. The image reconstruction method according to claim 2, wherein: The training image pair includes the first image and the second image obtained at the same energy.

4. The image reconstruction method according to claim 1, wherein: The training image pair includes the first image configured as a filtered back-projection reconstructed image and the second image configured as a filtered back-projection reconstructed image.

5. The image reconstruction method according to claim 1, wherein: The training image pair includes the first image configured as an iteratively reconstructed image and the second image configured as a filtered backprojection reconstructed image.

6. The image reconstruction method according to claim 1, wherein: The image data includes a reconstructed image, wherein an output image is calculated based on the reconstructed image using a trained neural network model.

7. The image reconstruction method according to claim 6, wherein: Acquisition of computed tomography image data includes: Acquisition of computed tomography data; Calculation is performed based on the computed tomography data to obtain the reconstructed image.

8. The image reconstruction method according to claim 6, wherein: The reconstructed image includes any one of the following two forms: A filtered back projection reconstructed image obtained based on a filtered back projection algorithm; and Iteratively reconstructed image obtained based on iterative reconstruction algorithm.

9. An image reconstruction device, comprising: an acquisition unit configured to acquire image data of a computer tomography scan; a processing unit configured to use a trained neural network model to perform processing based on the image data to obtain an output image, wherein the trained neural network model is configured to be trained via an image database including at least one training image pair consisting of a first image and a second image, and wherein the first image and the second image in the training image pair serve as a training input image and a training output image, respectively, and the first image and the second image are configured to be images of the same biological tissue obtained via computer tomography at the same time point; wherein the training image pair comprises the first image and the second image obtained via dual-source computed tomography; The dual-source computer tomography system includes a first probe and a second probe for simultaneously generating an X-ray beam for tomography, and a first receiver opposite to the first probe for receiving the X-ray beam emitted by the first probe and a first receiver opposite to the second probe for receiving the X-ray beam emitted by the second probe; During the dual-source computed tomography scanning process, the X-ray beam emitted by the first probe scans the biological tissue to be measured, and the first receiver receives the X-ray beam emitted by the first probe that passes through the biological tissue. At the same time, the second probe emits an X-ray beam to scan the biological tissue to be measured, and the second receiver receives the X-ray beam emitted by the second probe that passes through the biological tissue.

10. The image reconstruction device according to claim 9, wherein: The training image pair includes the first image and the second image acquired at different doses, wherein the dose when the first image is acquired is lower than the dose when the second image is acquired.

11. An electronic device, comprising: at least one processor; as well as a memory communicatively coupled to the at least one processor; in The memory stores a computer program, which, when executed by the at least one processor, implements the method according to any one of claims 1 to 8.

12. A non-transitory computer-readable storage medium storing a computer program, wherein: The computer program implements the method according to any one of claims 1 to 8 when executed by a processor.

13. A computer program product comprising a computer program, wherein: The computer program implements the method according to any one of claims 1 to 8 when executed by a processor.

Images