An image imaging method and system

Abstract

Embodiments of the present specification disclose an image imaging method and system. The method comprises: acquiring a first image of a target object and topological data, wherein the topological data comprises first topological data and second topological data, and the first topological data corresponds to the second topological data; determining a base material density image corresponding to the first image; determining a topological data difference based on the first topological data and the second topological data; and determining a second image corresponding to the first image based on the first image, the base material density image and the topological data difference.

Description

TECHNICAL FIELD

[0001] The present specification relates to the field of image processing, and in particular, to an image imaging method and system. BACKGROUND

[0002] When a CT (Computed Tomography) machine scans, an X-ray is emitted from a radiation source to a target object, a detector receives an X-ray attenuation signal passing through the target object, and a tomographic image of the target object is reconstructed by a computer, which has the characteristics of fast scanning time and clear image, and can be used for the examination of various diseases.

[0003] Dual-energy (DE) CT, there are several methods to perform dual-energy CT acquisition, such as dual source, fast voltage (kVp) switching and double-layer detector configuration, dual-energy CT has higher detection accuracy than traditional CT, which can accurately obtain the material information of the scanned object, it is a specific configuration of spectral CT, which uses two attenuation values collected at two different energy spectra to solve the photoelectric and Compton contributions, which includes the mass attenuation coefficient of the material, and thereby identifies unknown materials by their photoelectric and Compton contributions. Because of the photoelectric / Compton properties, iodine can be distinguished from, for example, calcium and water. Because any two linearly independent sums of the two basis functions span the entire attenuation coefficient space, any material can be represented by a linear combination of the other two materials (so-called basis materials) (e.g., water and iodine). New applications are provided, such as monochromatic images, material elimination images, effective atomic number images, and electron density images. Material decomposition of the second image or the first image or the image obtained by the traditional scan plays a significant role in many applications such as automatic separation of bone and contrast agent in enhanced scans, iodine quantification, qualitative analysis of kidney stones and other lesions, and generation of pseudo-single-energy images and virtual non-enhanced images. However, the existing dual-energy CT imaging method has some inconvenience in actual application, for example, the scanning time of dual-energy CT is longer, which will bring a higher radiation dose to the target object.

[0004] Therefore, it is necessary to propose an image imaging method and system to reduce the radiation dose to the target object when obtaining a high-energy CT image. SUMMARY

[0005] One aspect of the embodiments of the present specification provides an image imaging method. The image imaging method comprises: acquiring a first image of a target object and topological data, wherein the topological data comprises first topological data and second topological data, and the first topological data corresponds to the second topological data; processing the first image by using a preset image processing model to determine a base material density image corresponding to the first image; determining a topological data difference based on the first topological data and the second topological data; and determining a second image corresponding to the first image based on the first image, the base material density image, and the topological data difference.

[0006] In some embodiments, the determining the second image corresponding to the first image based on the first image, the base material density image, and the topological data difference comprises: determining an image decomposition matrix difference based on the base material density image and the topological data difference; determining an image difference based on the base material density image and the image decomposition matrix difference; and determining the second image based on the first image and the image difference.

[0007] In some embodiments, the determining the image decomposition matrix difference based on the base material density image corresponding to the first image and the topological data difference comprises: determining material density data based on the base material density image; and determining the image decomposition matrix difference based on the material density data and the topological data difference.

[0008] In some embodiments, the acquiring the first topological data comprises: acquiring the first topological data from scanning data corresponding to the first image; or performing scanning on the target object under the first radiation dose. In some embodiments, the determining the base material density image corresponding to the first image comprises: inputting the first image into a pre-trained image processing model to determine the base material density image corresponding to the first image.

[0009] In some embodiments, the image processing model is trained in the following manner: acquiring a plurality of training samples, wherein each training sample in the plurality of training samples comprises a sample first image; inputting the training sample into an image processing model to determine a sample base material density image corresponding to the sample first image; adjusting parameters of the image processing model based on one or more of a label base material density image, the sample first image, a sample second image, sample topological data, and the sample base material density image, to obtain a trained image processing model, wherein the sample second image corresponds to the sample first image.

[0010] In some embodiments, the target loss function comprises any one or combination of a base material density image loss function, a first image loss function, a second image loss function, and a third image loss function; wherein the base material density image loss function is determined based at least on the label base material density image, the first image loss function is determined based at least on the sample first image, the second image loss function is determined based at least on the sample second image, and the third image loss function is determined based on the sample second image and sample topology data; wherein the sample topology data comprises sample first topology data and sample second topology data, and the sample first topology data and the sample second topology data correspond.

[0011] In some embodiments, the first image is a low-energy image, and the second image is a high-energy image; wherein the low-energy image is obtained by scanning the target object under a first radiation dose, and the high-energy image is obtained by scanning the target object under a second radiation dose; the first radiation dose is lower than the second radiation dose.

[0012] Another aspect of the embodiments of the present specification provides an image imaging system. The system comprises: an acquisition module configured to acquire a first image of a target object and topology data, wherein the topology data comprises first topology data and second topology data, and the first topology data corresponds to the second topology data; a first determination module configured to determine a base material density image corresponding to the first image by processing the first image using a preset image processing model; a second determination module configured to determine a topology data difference based on the first topology data and the second topology data; and a third determination module configured to determine a second image corresponding to the first image based on the first image, the base material density image, and the topology data difference.

[0013] Another aspect of the embodiments of the present specification provides an image imaging apparatus comprising at least one storage medium and at least one processor, the at least one storage medium being configured to store computer instructions; and the at least one processor being configured to execute the computer instructions to implement an image imaging method.

[0014] Another aspect of the embodiments of the present specification provides a computer readable storage medium, the storage medium storing computer instructions, when a computer reads the computer instructions in the storage medium, the computer executes an image imaging method.

[0015] In the embodiments of the present specification, the trained image processing model can be used to obtain the corresponding second image based on the single-energy first image, which can reduce the radiation dose caused by scanning the target object using rays. At the same time, by introducing topological data, more information can be obtained in the calculation process, which can improve the accuracy of the dual-energy image post-processing result; before the first image is input into the image processing model, the first image can be processed without denoising and the like, which can simplify the process of obtaining the base material density image; and using the image processing model to obtain the base material density image can avoid or reduce the problem of serious degradation of image signal-to-noise ratio caused by using matrix inversion on the dual-energy image, and save the time spent on iterative material decomposition. BRIEF DESCRIPTION OF DRAWINGS

[0016] The present specification will be further illustrated in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same numbers represent the same structures, in which:

[0017] Figure 1 is a schematic diagram of an exemplary application scenario of an image imaging system according to some embodiments of the present specification;

[0018] Figure 2 is an exemplary flowchart of an image imaging method according to some embodiments of the present specification;

[0019] Figure 3 is an exemplary flowchart of determining a second image according to some embodiments of the present specification;

[0020] Figure 4 is an exemplary flowchart of a training method of an image processing model according to some embodiments of the present specification;

[0021] Figure 5 is an exemplary flowchart of determining a base material density image loss function according to some embodiments of the present specification;

[0022] Figure 6 is an exemplary flowchart of determining a second image loss function according to some embodiments of the present specification;

[0023] Figure 7 is an exemplary flowchart of determining a third image loss function according to some embodiments of the present specification;

[0024] Figure 8 is an exemplary flowchart of determining a topological second image according to some embodiments of the present specification;

[0025] Figure 9 is an exemplary module diagram of an image imaging system according to some embodiments of the present specification. DETAILED DESCRIPTION

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings required to be used in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can also be applied to other similar scenarios without creative labor on the basis of these drawings. Unless it is clear from the language environment or otherwise stated, the same reference numbers in the drawings represent the same structure or operation.

[0027] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0028] As shown in the specification and claims, unless the context clearly indicates otherwise, "a", "an", "one", and / or "this" do not refer to the singular, but can also include the plural. Generally, the terms "comprise" and "include" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0029] Flowcharts are used in the present specification to illustrate the operations performed by the system according to the embodiments of the present specification. It should be understood that the preceding or subsequent operations are not necessarily performed in sequence. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or more steps of the operation can be removed from these processes.

[0030] Computed Tomography (CT) is currently widely used in clinical diagnosis. Dual-energy CT technology has developed rapidly and has gradually become one of the common examination methods in clinical diagnosis.

[0031] Currently, one implementation of clinical dual-energy CT imaging is achieved through hardware design, for example, kVp switching technology, double-layer detector technology, dual-source CT imaging technology, and multiple scanning technology, etc., and then direct matrix inversion or iterative material decomposition in the image domain is performed to obtain the base material density image. The fast kVp switching technology only uses a single radiation source, and a single tube can be used to achieve high and low voltage switching to acquire dual-energy data. The scanning time is usually long, the differentiation ability between different energy levels is poor, the difference between the generated second image and the first image is small, and an additional scan is also required. The dual-source CT imaging technology uses two radiation sources to emit X-rays of two energy levels. The two radiation sources are at a certain angle about the center of rotation. The two radiation sources can emit X-rays at the same time, and the scanning range of low-energy X-rays and high-energy X-rays at the same time is quite different. Although the dual-source CT imaging technology has good energy level differentiation ability, the hardware implementation is complex, and due to the existence of a certain angle offset between the two radiation sources, the generated dual-energy images in the motion-intensive area are prone to offset or distortion, and may therefore produce artifacts in the images obtained by material decomposition. In addition, the dual-energy CT technologies shown in the above examples all have a common problem that the radiation dose to the patient is large.

[0032] The embodiments of the present specification disclose a training method of an image processing model. The model is trained by combining machine learning technology. The second image is obtained by combining the single-energy CT image and the topological data through the trained image processing model, so as to reduce the radiation dose to the target object when obtaining the second image. The technical solutions disclosed in the present specification are described in detail through the description of the drawings.

[0033] Figure 1 is an example application scenario of an image imaging system according to some embodiments of the present specification.

[0034] In some embodiments, the image processing system 100 can be used to obtain a high-energy CT image based on a low-energy CT image, or to obtain a low-energy CT image based on a high-energy CT image.

[0035] For example, in a typical application scenario, the image processing system 100 can acquire a first image and topological data of a target object, wherein the topological data includes first topological data and second topological data, and the first topological data corresponds to the second topological data. The image processing system 100 can determine the matrix material density image corresponding to the first image. The image processing system 100 can determine the topological data differences based on the first topological data and the second topological data. The image processing system 100 can determine the second image corresponding to the first image based on the first image, the matrix material density image, and the topological data differences. In some embodiments, the first image can be a low-energy CT image, and the second image can be a high-energy CT image. In some embodiments, the first image can be a high-energy CT image, and the second image can be a low-energy CT image.

[0036] It should be noted that, in this specification, dual-energy images refer to a high-energy CT image and its corresponding low-energy CT image. When the first image is a low-energy CT image, the corresponding second image is a high-energy CT image; when the first image is a high-energy CT image, the corresponding second image is a low-energy image.

[0037] In some embodiments, the image processing model can be trained in other systems (e.g., a training system for the image processing model, not shown) or using the image processing system 100. For example, the image processing system 100 acquires multiple training samples, each of which includes a first sample image; inputs the training samples into the image processing model to determine the base material density image corresponding to the first sample image; and adjusts the parameters of the image processing model based on one or more of the labeled base material density image, the first sample image, the second sample image, sample topology data, and the base material density image, with the optimization of the target loss function as the training objective, to obtain a trained image processing model; wherein the second sample image corresponds to the first sample image.

[0038] like Figure 1 As shown, the image processing system 100 may include an imaging device 110, a network 120, a terminal 130, a processing device 140, and a storage device 150.

[0039] The imaging device 110 can be used to image a target object to generate an image. The imaging device 110 can be a medical imaging device (e.g., a CT (Computed Tomography). In some embodiments, the imaging device 110 can include a gantry 111, a detector 112, a scan region 113, and a scan bed 114. A target object can be placed on the scan bed 114 to receive a scan. The gantry 111 can support the detector 112. In some embodiments, the detector 112 can include one or more detector units. The detector units can be and / or include single-row detectors and / or multi-row detectors. The detector units can include scintillation detectors (e.g., cesium iodide detectors) and other detectors, etc. In some embodiments, the gantry 111 can rotate, e.g., in a CT imaging device, the gantry 111 can rotate clockwise or counterclockwise around a gantry rotation axis. In some embodiments, the imaging device 110 can further include a radiation scan source, which can rotate with the gantry 111. The radiation scan source can emit a radiation beam (e.g., X-rays) to the target object, which is attenuated by the target object and detected by the detector 112 to generate an image signal. In some embodiments, the scan bed 114 can be movably disposed in front of the machine and parallel to the ground. The scan bed 114 can be moved to enter and exit the scan region 113. The scan bed 114 can also be moved in a vertical direction to adjust the distance between the target object on the scan bed and the detector 112 (or the scan center) when entering the scan region 113, so as to scan the target object in the scan range.

[0040] The processing device 140 can process data and / or information obtained from the imaging device 110, the terminal 130, and / or the storage device 150. For example, the processing device 140 can process image information detected by the detector 112 to obtain a CT image. For another example, the processing device 140 can process a low-energy CT image to obtain a high-energy CT image corresponding to the low-energy CT image. In some embodiments, the processing device 140 can be a single server or a group of servers. The group of servers can be centralized or distributed. In some embodiments, the processing device 140 can be local or remote. For example, the processing device 140 can access information and / or data from the imaging device 110, the terminal 130, and / or the storage device 150 through the network 120. For another example, the processing device 140 can directly connect to the imaging device 110, the terminal 130, and / or the storage device 150 to access information and / or data. In some embodiments, the processing device 140 can be implemented on a cloud platform. For example, the cloud platform can include one or a combination of a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, a cross-cloud, a multi-cloud, etc.

[0041] The terminal 130 can include a mobile device 131, a tablet 132, a notebook 133, or the like, or any combination thereof. In some embodiments, the terminal 130 can interact with other components in the image processing system 100 through a network. For example, the terminal 130 can send one or more control instructions to the imaging device 110 to control the imaging device 110 to scan the target object according to the instructions. For another example, the terminal 130 can also receive the base material density image determined by the processing device 140 and display the base material density image for an operator to analyze and confirm. In some embodiments, the mobile device 131 can include a smart home device, a wearable device, a mobile device, a virtual reality device, an augmented reality device, or the like, or any combination thereof. In some embodiments, the smart home device can include a smart lighting device, a smart appliance control device, a smart monitoring device, a smart television, a smart radiograph, an intercom, or the like, or any combination thereof. In some embodiments, the wearable device can include a bracelet, a shoe, a pair of glasses, a helmet, a watch, a piece of clothing, a backpack, a smart accessory, or the like, or any combination thereof. In some embodiments, the mobile device can include a mobile phone, a personal digital assistant (PDA), a gaming device, a navigation device, a POS device, a notebook, a tablet, a desktop, or the like, or any combination thereof. In some embodiments, the virtual reality device and / or the augmented reality device can include a virtual reality helmet, a virtual reality pair of glasses, a virtual reality patch, an augmented reality helmet, an augmented reality pair of glasses, an augmented reality patch, or the like, or any combination thereof. For example, the virtual reality device and / or the augmented reality device can include or and the like. In some embodiments, the terminal 130 can be a part of the processing device 140. In some embodiments, the terminal 130 can be integrated with the processing device 140 as an operation console of the imaging device 110. For example, a user / operator (e.g., a doctor) of the image processing system 100 can control the operation of the imaging device 110, such as scanning the target object, controlling the movement of the scanning bed 114, training the image processing model, obtaining the base material density image, obtaining the second image based on the first image and the topological data, and the like, through the operation console.

[0042] The storage device 150 can store data (e.g., scan data of a target object), instructions, and / or any other information. In some embodiments, the storage device 150 can store data obtained from the imaging device 110, the terminal 130, and / or the processing device 140, e.g., the storage device 150 can store a treatment plan obtained from the imaging device 110, scan data of a target object, etc. In some embodiments, the storage device 150 can store data and / or instructions that the processing device 140 can execute or use to perform the exemplary methods described in the present application. In some embodiments, the storage device 150 can include one or a combination of a mass storage, a removable storage, a volatile read / write memory, a read-only memory (ROM), etc. The mass storage can include a disk, an optical disk, a solid state disk, a mobile storage, etc. The removable storage can include a flash drive, a floppy disk, an optical disk, a memory card, a ZIP disk, a magnetic tape, etc. The volatile read / write memory can include a random access memory (RAM). The RAM can include a dynamic random access memory (DRAM), a double data rate synchronous dynamic random access memory (DDR-SDRAM), a static random access memory (SRAM), a thyristor random access memory (T-RAM), a zero-capacitor random access memory (Z-RAM), etc. The ROM can include a mask read-only memory (MROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a compact disk read-only memory (CD-ROM), an optical disk of a digital versatile disk, etc. In some embodiments, the storage device 150 can be implemented by a cloud platform described in the present application. For example, the cloud platform can include one or a combination of a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, a cross-cloud, a multi-cloud, etc.

[0043] In some embodiments, the storage device 150 can be connected to the network 120 to enable communication between one or more components in the image processing system 100 (e.g., the processing device 140, the terminal 130, etc.). One or more components in the image processing system 100 can read data or instructions in the storage device 150 through the network 120. In some embodiments, the storage device 150 can be a part of the processing device 140 or can be independent and directly or indirectly connected to the processing device.

[0044] The network 120 can include any suitable network capable of facilitating the exchange of information and / or data between components of the image processing system 100. In some embodiments, one or more components of the image processing system 100 (e.g., the imaging device 110, the terminal 130, the processing device 140, the storage device 150, etc.) can exchange information and / or data between one or more components of the image processing system 100 via the network 120. For example, the processing device 140 can obtain scan data from the imaging device 110 via the network 120. The network 120 can include one or a combination of public networks (e.g., the Internet), private networks (e.g., local area networks (LANs), wide area networks (WANs)), wired networks (e.g., Ethernet networks), wireless networks (e.g., 802.11 networks, wireless Wi-Fi networks, etc.), cellular networks (e.g., long term evolution (LTE) networks), frame relay networks, virtual private networks (VPNs), satellite networks, telephone networks (e.g., voice and / or data networks), routers, hubs, server computers, and / or the like. For example, the network 120 can include a wired network, a fiber optic network, a telecommunications network, a local area network, a wireless local area network (WLAN), a metropolitan area network (MAN), a wide area network (WAN), a private network, the Internet, the Public Switched Telephone Network (PSTN), a cable network, a public land mobile network (PLMN), a satellite network, a cellular network, a near field communication (NFC) network, and / or the like. In some embodiments, the network 120 can include one or more network access points. For example, the network 120 can include wired and / or wireless network access points, such as base stations and / or Internet exchange points, through which one or more components of the image processing system 100 can connect to the network 120 to exchange data and / or information. The network, The network, a near field communication (NFC) network, and / or the like. In some embodiments, the network 120 can include one or more network access points. For example, the network 120 can include wired and / or wireless network access points, such as base stations and / or Internet exchange points, through which one or more components of the image processing system 100 can connect to the network 120 to exchange data and / or information.

[0045] Figure 2 is an exemplary flowchart of an image imaging method according to some embodiments of the present specification. In some embodiments, the flow 200 can be performed by a processing device. For example, the flow 200 can be stored in a storage device (such as a self-provided storage unit of the processing device or an external storage device) in the form of a program or instruction, which when executed, can implement the flow 200. The flow 200 can include the following operations.

[0046] It should be noted that in the embodiments of the present specification, the first image is taken as the low-energy image and the second image is taken as the high-energy image for the main description, but this is only for the purpose of example. It should be understood that the first image can also be a high-energy image and the second image can be a low-energy image, which is not limited by the present specification.

[0047] In some embodiments, the low-energy image is obtained by scanning the target object at a first radiation dose, and the high-energy image is obtained by scanning the target object at a second radiation dose. At this time, the first radiation dose is lower than the second radiation dose.

[0048] At step 202, a first image and topological data of a target object are acquired. In some embodiments, step 202 can be performed by the acquisition module 910.

[0049] The target object can include a patient or other medical experimental subject (e.g., a laboratory mouse or other animal), etc. The target object can also be a part of a patient or other medical experimental subject, including organs and / or tissues, such as a heart, a lung, a rib, an abdominal cavity, etc.

[0050] In some embodiments, the first image can refer to a low-energy CT image obtained by scanning the target object with a low dose of radiation. In some embodiments, the first image can refer to a high-energy CT image obtained by scanning the target object with a high dose of radiation.

[0051] The topological data refers to scanning data of the target object at a certain angle (which can be any angle, such as a transverse plane, a coronal plane, a sagittal plane, etc.). The topological data includes first topological data and second topological data. In some embodiments, the first topological data can refer to topological data scanned with a lower energy level of radiation, and the second topological data can refer to topological data scanned with a higher energy level of radiation. In some embodiments, the first topological data can also refer to topological data scanned with a higher energy level of radiation, and the second topological data can refer to topological data scanned with a lower energy level of radiation. The first topological data and the second topological data correspond to each other, for example, both are scanning data of the same target object. In some embodiments, the first topological data and the second topological data can be scanned at the same angle. In some embodiments, the scanning angle of the topological data and the first image can be different, so that more information can be included in the topological data and the first image.

[0052] The first image and the first topological data are obtained by scanning the target object at a first radiation dose, and the second topological data is obtained by scanning the target object at a second radiation dose. For example, the imaging device can scan the target object using radiation (e.g., X-rays, etc.) with a lower energy (e.g., a first radiation dose) to obtain the first image and the first topological data. For example, the first topological data can be obtained from the scanning data corresponding to the first image; or the target object can be scanned at the first radiation dose. The imaging device can scan using radiation with a higher energy level (e.g., a second radiation dose) to obtain the second topological data. The first radiation dose and the second radiation dose are different.

[0053] In some embodiments, the processing device can obtain the first image and the topological data of the target object by reading from the imaging device, a database, a storage device, or by calling a relevant data interface, or by scanning using the imaging device.

[0054] In step 204, the base material density image corresponding to the first image is determined. In some embodiments, step 204 can be performed by the first determination module 920.

[0055] The base material density image refers to an image that can reflect the material decomposition result of the scanned object through density.

[0056] The essence of CT imaging is attenuation coefficient imaging. The principle is that X-rays have penetrating properties and can penetrate objects (e.g., living beings, objects). Different tissues of the object have different absorption and transmission rates of X-rays. When X-rays pass through the object, the attenuation of different parts is different. After measuring the attenuated rays, the data of different parts of the object can be obtained. After inputting the obtained data into an electronic computer for processing, the cross-section or three-dimensional image of the inspected part of the object can be reconstructed. The linear attenuation coefficient of any object can be represented as a linear combination of selected materials. Since the CT image has a linear relationship with the linear attenuation coefficient, the CT image of any object can be represented as a linear combination of selected materials. For example, the CT image can be represented by equation (1).

[0057] I = M · A (1)

[0058] Where I represents a CT image, with a dimension of N x 1, N being the total number of pixels in a CT image; M is the base material density image, with a dimension of N x m, m being the number of decomposed materials; and A is the material decomposition matrix, with a dimension of m x 1.

[0059] In some embodiments, the processing device can input the first image into a preset image processing model, and output the base material density image corresponding to the first image from the image processing model. In some embodiments, the image processing model can include a U-net model, a V-net model, etc. based on deep learning.

[0060] The preset image processing model can be a pre-trained image processing model. For example, the processing device can obtain a plurality of training samples, wherein each training sample in the plurality of training samples includes a sample first image. The training sample is input into the image processing model to determine the base material density image corresponding to the sample first image. Based on one or more of the label base material density image, the sample first image, the sample second image, and the sample topological data, and the base material density image, the parameters of the image processing model are adjusted to obtain a trained image processing model, wherein the sample second image corresponds to the sample first image. For more details about the training of the image processing model, please refer to the related description of Figures 4 to 8 .

[0061] At step 206, the processing device determines a topological data difference based on the first topological data and the second topological data. In some embodiments, the step 206 can be performed by the second determining module 930.

[0062] The topological data difference refers to a difference between the energy topological data of two different energy levels.

[0063] In some embodiments, the processing device can subtract the first topological data from the second topological data to obtain the topological data difference. For example, the high-energy topological data is subtracted from the low-energy topological data, or the low-energy topological data is subtracted from the high-energy topological data.

[0064] At step 208, the processing device determines a second image corresponding to the first image based on the first image, the base material density image, and the topological data difference. In some embodiments, the step 208 can be performed by the third determining module 940.

[0065] The second image is a second image obtained after a series of processing based on the first image and the topological data. In some embodiments, the second image can be a high-energy image. In some embodiments, the second image can be a low-energy image.

[0066] In some embodiments, the processing device can determine an image decomposition matrix difference based on the base material density image and the topological data difference. Then, the processing device determines an image difference based on the base material density image and the image decomposition matrix difference, and finally determines the second image based on the first image and the image difference. Details of determining the second image can be found in the description of Figure 3 and related descriptions, which will not be repeated here.

[0067] In the embodiments of the present disclosure, the second image corresponding to the first image can be obtained based on the base material density image and the topological data. Since the imaging device usually needs to scan the target object from multiple angles when obtaining the dual-energy image, the scanning time is relatively long, which causes a high radiation dose to the target object. However, the topological data can be only the scanning data of one angle of the target object, and only one angle of scanning only needs a very short time of scanning of the target object, thereby effectively reducing the radiation dose to the target object. In addition, compared with the way of directly obtaining the second image based on the first image, in the embodiments, by introducing the topological data, more information can be obtained in the calculation process, thereby improving the dual-energy CT value and improving the accuracy of the dual-energy image post-processing result.

[0068] Figure 3is an exemplary flowchart of determining a second image according to some embodiments of the present specification. In some embodiments, the flow 300 can be performed by a processing device. For example, the flow 300 can be stored in a storage device (such as a self-storage unit of the processing device or an external storage device) in the form of a program or instructions which, when executed, can implement the flow 300. As shown, the flow 300 can include the following operations. Figure 3

[0069] At step 302, an image decomposition matrix difference is determined based on the base material density image and the topological data difference.

[0070] The image decomposition matrix difference refers to a difference between an image decomposition matrix of the first image and an image decomposition matrix of the second image.

[0071] In some embodiments, the image decomposition matrix difference can be determined based on the material density data and the topological data difference. The material density data refers to data obtained by forward projecting the base material density image.

[0072] Exemplarily, for a dual-energy CT image, it can be represented by the following equations (2) and (3), respectively.

[0073] I Low = M·A Low (2)

[0074] I High = M·A High (3)

[0075] where I Low represents the first image, I High represents the second image, A Low is a material decomposition matrix of the first image, A High is a material decomposition matrix of the second image, and M is a base material density image.

[0076] Subtracting the first image from the second image can obtain equation (4).

[0077] I diff = M·A diff (4)

[0078] where I diff represents an image difference between the second image and the first image, and A diff represents a decomposition matrix difference between the second image and the first image.

[0079] Simultaneously forward projecting both sides of equation (4), the symbol of forward projection is denoted as R, then equation (5) can be obtained after forward projection.

[0080] R·I​diff = R MA diff (5)

[0081] wherein R I diff represents the energy data difference between the second image and the first image; R M represents the material density data; and A diff represents the decomposition matrix difference between the second image and the first image. In some embodiments, A diff may be solved based on equation (4) by using a least square fitting method.

[0082] Based on the above description, the processing device can determine the image decomposition matrix difference based on the process shown in the following example.

[0083] In some embodiments, the processing device can determine the material density data based on the base material density image. For example, the processing device can forward project the base material density image to obtain the material density data.

[0084] The processing device can determine the image decomposition matrix difference based on the material density data and the topology data difference.

[0085] For example, the processing device can subtract the first topology data from the second topology data to obtain the topology data difference. The processing device can obtain the image decomposition matrix difference based on equation (5) described above, wherein the topology data difference is R I diff , which represents the energy difference between the second topology data and the first topology data, the material density data can be obtained by forward projecting the base material density image, i.e. R M, and finally the image decomposition matrix difference A diff can be obtained by least square fitting calculation.

[0086] Step 304, determine the image difference based on the base material density image and the image decomposition matrix difference.

[0087] The image difference refers to the difference between the second image and the first image. After obtaining the image decomposition matrix difference based on the topology data difference and the base material density image, the image difference can be obtained by equation (4) shown in the above example. In equation (4), M is the base material density image, A diff is the obtained image decomposition matrix difference, and the image difference I diff can be obtained by substituting them into equation (4).

[0088] Step 306, determine the second image based on the first image and the image difference.

[0089] In some embodiments, the processing device can determine the second image based on the first image and the image difference, and equation (3) and equation (4). Since Idiff = I High - I Low It has been found that M·A diff It has also been found that I Low If the first image is I, then the second image can be obtained by adding the image difference to the first image.

[0090] Figure 4 is an exemplary flowchart of a method of training an image processing model according to some embodiments of the present specification. In some embodiments, the flow 400 can be performed by a processing device (e.g., the processing device 140). For example, the flow 400 can be stored in a storage device (such as a self-contained storage unit of the processing device or an external storage device) in the form of a program or instructions that, when executed, can implement the flow 400. The flow 400 can include the following operations.

[0091] At step 402, a plurality of training samples is obtained.

[0092] The training samples include low-energy image data used to train the image processing model.

[0093] In some embodiments, each training sample in the plurality of training samples can include a sample low-energy image. A low-energy image refers to an image obtained by scanning a target object with a low dose of radiation. The target object can include a patient, or other medical experimental subjects (e.g., laboratory mice or other animals), etc. The target object can also be a part of a patient or other medical experimental subject, including organs and / or tissues, such as the heart, lungs, ribs, abdominal cavity, etc.

[0094] In some embodiments, the processing device can obtain the plurality of training samples by reading from a database, a storage device, or an imaging device.

[0095] At step 404, the training samples are input to the image processing model to determine a sample base material density image corresponding to the sample first image.

[0096] In some embodiments, the processing device can input the training samples (sample first images) to the image processing model, and output a sample base material density image corresponding to the sample first image by the image processing model. The sample base material density image refers to a base material density image output by the image processing model during training of the model.

[0097] The base material density image output by the image processing model can be used to constrain the training of the image processing model, or can be used to obtain a second image corresponding to the first image. The related description of obtaining the second image can be referred to other parts of the present specification, for example, the related description of Figures 2 to 3 , which will not be repeated here.

[0098] At step 406, parameters of the image processing model are adjusted based on one or more of the label base material density image, the sample first image, the sample second image, the sample topological data, and the sample base material density image, with an optimization target loss function as a training target, to obtain a trained image processing model.

[0099] The sample second image corresponds to the sample first image. For example, the sample second image and the sample first image are obtained by scanning the same target object.

[0100] In some embodiments, the processing device can input a training sample into an initial image processing model, and based on a prediction result (i.e., an output sample base material density image) of the image processing model, and based on one or more of the label base material density image, the sample first image, the sample second image, and the sample topological data, a loss function is constructed to constrain, by continuously adjusting parameters of the image processing model, to optimize a loss function value corresponding to each training sample, for example, to minimize the loss function value, so that the prediction result of the final image processing model can be more accurate. When the loss function value meets the requirement (e.g., less than a certain preset value, the loss function value converges) or the iteration reaches a preset number of times, a trained image processing model can be obtained.

[0101] In some embodiments, the target loss function corresponding to each training sample can be any one or combination of the base material density image loss function, the first image loss function, the second image loss function, and the third image loss function. For example, the target loss function can be a combination of the base material density image loss function and the first image loss function, or a combination of the base material density image loss function, the first image loss function, the second image loss function, and the third image loss function.

[0102] In some embodiments, the base material density image loss function can be determined based on at least the label base material density image. For example, the base material density image loss function can be constructed based on the sample base material density image predicted by the image processing model and the label base material density image. Exemplarily, the base material density image loss function can be as shown in equation (6).

[0103]

[0104] wherein L Maerial denotes the base material density image loss function; I Low is the sample first image, F(I Low ) is the sample base material density image predicted by the image processing model; Material is the label base material density image; and N is the total number of training samples used for model training.

[0105] In this embodiment, the loss function is constructed directly based on the label basis material density image, which can constrain the trained image processing model, so that the prediction result of the model is close to the label basis material density image obtained based on the dual-energy CT image. The first image is input into the trained image processing model to obtain the basis material density image, thereby realizing the material decomposition function using a single-energy CT image. Since the basis material density image can be obtained using only a single-energy CT image, compared with using a dual-energy CT image to obtain the basis material density image, the scanning time of the target object is shorter, and the radiation dose of the target object is lower.

[0106] In some embodiments, the first image loss function can be determined based at least on the sample first image. For example, the first image loss function can be determined based on the sample first image and the predicted first image. In some embodiments, the predicted first image can be calculated based on the basis material density image and the sample first image. For example, the predicted first image can be obtained by multiplying the basis material density image by a material decomposition matrix corresponding to the sample first image. The material decomposition matrix of the sample first image can be obtained by solving the sample first image and the basis material density image by least square fitting, and more details can be referred to the related description of Figure 4 , which will not be repeated here.

[0107] For example, the first image loss function can be represented by equation (7).

[0108]

[0109] wherein, is the predicted first image; I Low is the sample first image; and N is the total number of training samples. The value of the first image loss function can reflect the difference between the predicted first image and the sample first image, and the predicted first image is solved based on the basis material density image, so it can indirectly reflect the accuracy of the prediction result of the model.

[0110] In some embodiments, the second image loss function can be determined based at least on the sample second image. For example, the second image loss function can be constructed based on the sample second image and the predicted second image. In some embodiments, the predicted second image can be calculated based on the basis material density image and the sample second image. For example, the predicted second image can be obtained by multiplying the basis material density image by a material decomposition matrix corresponding to the sample second image. The material decomposition matrix of the sample second image can be obtained by solving the sample second image and the basis material density image by least square fitting, and more details can be referred to the related description of Figure 4 , which will not be repeated here.

[0111] Exemplarily, the second image loss function L High-1 may be shown as equation (8).

[0112]

[0113] wherein, is a predicted second image; I High is a sample second image; and N is a total number of training samples.

[0114] In this embodiment, the predicted second image is calculated by the predicted base material density image and the sample second image, and then the image processing model is constrained in the manner of constructing the loss function based on the predicted second image and the sample second image. Since the predicted second image is calculated based on the predicted base material density image, the loss function constructed by using the predicted second image and the sample second image can reflect the accuracy of the predicted base material density image to some extent. Through the trained image processing model, the purpose of obtaining the base material density image using the single-energy first image can be achieved, and the single-energy CT image can be used to realize the material decomposition function.

[0115] In some embodiments, the third image loss function can be determined based at least on the sample second image and the sample topological data. For example, the third image loss function can be constructed based on the sample second image and a topological second image obtained according to the sample topological data. In some embodiments, the topological second image corresponds to the second image. For example, the second image is a high-energy image, and the topological second image is a topological high-energy image. The second image is a low-energy image, and the topological second image is a topological low-energy image.

[0116] Details about determining the topological second image based on the sample topological data can be referred to Figure 6 and related descriptions thereof, which will not be repeated here.

[0117] In some embodiments, the third image loss function L High-2 may be shown as equation (9).

[0118]

[0119] wherein, is a topological second image; I High is a sample second image; and N is a total number of training samples.

[0120] In this embodiment, the topology second image is obtained by the predicted base material density image and the sample topology data, and then the model is constrained in the manner of constructing the loss function based on the topology second image and the sample second image. Since the topology second image is solved based on the predicted base material density image and the sample topology data, the loss function constructed using the topology second image and the sample second image can also reflect the accuracy of the predicted base material density image to some extent.

[0121] Exemplarily, the target loss function of the combination of part loss functions is exemplarily shown in the following embodiments, for example, the combination of the base material density image loss function, the first image loss function and the second image loss function can be shown as equation (10).

[0122] L1=λ·L Maerial +L Low +L High-1 (10)

[0123] Wherein, L1 represents the combination loss function of the base material density image loss function, the first image loss function and the second image loss function; λ is a weight balance factor, λ is a constant; L Maerial is the base material density image loss function, L Low represents the first image loss function; L High-1 represents the second image loss function.

[0124] In some embodiments, the target loss function of the combination of the base material density image loss function, the first image loss function, the second image loss function and the third image loss function can be shown as equation (11).

[0125] L2=λ·L Maerial +L Low +L High-1 +L High-2 (11)

[0126] Wherein, L2 represents the combination loss function of the base material density image loss, the first image loss function, the second image loss function and the third image loss function; λ is a weight balance factor, λ is a constant; L Maerial represents the base material density image loss function, L Low represents the first image loss function; L High-1 represents the second image loss function, L High-2 represents the third image loss function. For more detailed description of each part of the loss function, please refer to the relevant description above, which will not be repeated here.

[0127] Preferably, in some embodiments, the image processing model can be constrained by the combination loss function shown in equation (11), the constraint condition of the target loss function shown in equation (11) is more, and the trained model can have stronger robustness, so that the prediction result can be better.

[0128] By constraining the model by the loss function shown in the above examples, the trained image processing model can be obtained after the loss function value meets the preset condition (for example, the loss function value converges, is less than the preset threshold, etc.) or reaches the preset iteration number. The trained image processing model can be used to obtain the base material density image.

[0129] In embodiments, the image processing model is trained using training sample data, and the model can be constrained by a loss function constructed based on one or a combination of the labeled base material density image, the sample first image, the sample second image, and the sample topological data. The trained model can realize the material decomposition function by using a single-energy CT image. In the case of constraining the model by the combination loss function, the robustness of the model can be improved due to the addition of more constraint conditions, so that the prediction result of the trained model can be more accurate. At the same time, since the base material density image can be obtained only by using a single-energy CT image, compared with obtaining the base material density image using a dual-energy CT image, the single-energy CT image has a shorter scanning time and a lower radiation dose to the target object, compared with the dual-energy CT image which needs to scan the target object multiple times, has a longer scanning time, and has a higher radiation dose to the target object. In addition, by using the trained image processing model to process the image, compared with directly obtaining the base material density image by matrix inversion decomposition or iterative material decomposition, the base material density image signal-to-noise ratio will not be severely degraded, and the time required to obtain the base material density image can be reduced.

[0130] Figure 5 is an exemplary flowchart of determining the base material density image loss function according to some embodiments of the present specification. In some embodiments, flow 500 can be executed by a processing device. For example, flow 500 can be stored in the form of a program or instruction in a storage device (such as a self-provided storage unit of a processing device or an external storage device), which when executed, can implement flow 500. As shown, flow 500 can include the following operations. Figure 5

[0131] Step 502, processing the sample first image by the image processing model to obtain a sample base material density image corresponding to the sample first image.

[0132] ​In some embodiments, the processing device can input the sample first image to the image processing model, and obtain the sample base material density image by processing the sample first image by the image processing model.

[0133] At step 504, the base material density image loss function is determined based on the sample base material density image and the label base material density image.

[0134] In some embodiments, the determined base material density image loss function is shown in equation (6).

[0135] In some embodiments, the processing device can determine the difference between the sample base material density image and the label base material density image based on the base material density image loss function shown in step 204 (i.e., equation (6)). For example, the sample base material density image and the label base material density image are substituted into equation (6) to calculate the value of the base material density image loss function. The size of the value of the base material density image loss function can reflect the difference between the sample base material density image and the label base material density image. By minimizing the loss function, the difference can be reduced, so that the prediction result of the model is more accurate.

[0136] Figure 6 is an exemplary flowchart for determining the value of the first image loss function according to some embodiments of the present specification. In some embodiments, the flow 600 can be executed by the processing device. For example, the flow 600 can be stored in the form of programs or instructions in a storage device (such as a self-provided storage unit of the processing device or an external storage device), which when executed, can implement the flow 600. As shown, the flow 600 can include the following operations. Figure 6

[0137] At step 602, the sample first image is processed by the image processing model to obtain the sample base material density image corresponding to the sample first image.

[0138] The process of obtaining the sample base material density image of the sample first image is the same as the exemplary process described in Figure 5 , and more details can be referred to the related description of Figure 5 , which will not be described here.

[0139] At step 604, the predicted first image is determined based on the sample base material density image.

[0140] The predicted first image refers to the first image calculated based on the prediction result (i.e., the base material density image) of the image processing model base material density image.

[0141] In some embodiments, the predicted first image can be determined based on the following equation (12).​

[0142]

[0143] wherein, represents the predicted first image; M is the predicted sample basis material density image; A low is the material decomposition matrix of the sample first image. The sample basis material density image M is multiplied by the material decomposition matrix A low , and the predicted first image A low The sample first image and the sample basis material density image can be solved by a least squares fitting method.

[0144] Step 606, determining the first image loss function based on the predicted first image and the sample first image.

[0145] In some embodiments, the first image loss function can be calculated as equation (7) in step 406 above. The processing device can substitute the predicted first image and the sample first image into equation (7) to obtain the value of the first image loss function.

[0146] Step 608, determining the predicted second image based on the sample basis material density image.

[0147] The predicted second image refers to a second image calculated based on the predicted result (i.e. basis material density image) of the image processing model basis material density image.

[0148] In some embodiments, the predicted second image can be determined based on the following equation (13).

[0149]

[0150] wherein, represents the predicted second image; M is the predicted sample basis material density image; A High is the material decomposition matrix of the sample second image. The sample basis material density image M is multiplied by the material decomposition matrix A High , and the predicted second image

[0151] In some embodiments, A High The sample second image and the sample basis material density image can be solved by a least squares fitting method.

[0152] Step 610, determining the sample second image loss function based on the predicted second image and the sample second image.

[0153] In some embodiments, the second image loss function can be as shown in step 406 above. The processing device can substitute the predicted second image and the sample second image into equation (8) to obtain the value of the sample second image loss function.

[0154] Figure 7 is an exemplary flowchart illustrating determining a third image loss function according to some embodiments of the present disclosure. In some embodiments, the flow 700 can be performed by a processing device. For example, the flow 700 can be stored in a storage device (such as a self-provided storage unit of the processing device or an external storage device) in the form of a program or instructions, which when executed, can implement the flow 700. As shown in Figure 7 the flow 700 can include the following operations.

[0155] At step 702, a topological second image is determined based on the base material density image, the sample first image, and the sample topological data.

[0156] The topological second image refers to a second image calculated based on the predicted result (i.e., the sample base material density image) of the image processing model, the sample first image, and the sample topological data.

[0157] The sample topological data includes sample low-energy topological data and sample high-energy topological data.

[0158] Details regarding determining the topological second image can be found in Figure 8 and related descriptions thereof, which will not be repeated here.

[0159] At step 704, the third high-energy image loss function is determined based on the sample second image and the topological second image.

[0160] In some embodiments, the determined third high-energy image loss function can be as shown in equation (9) of step 406.

[0161] In some embodiments, the processing device can substitute the topological second image and the sample second image into the third high-energy image loss function to obtain the value of the third high-energy image loss function.

[0162] Figure 8 is an exemplary flowchart illustrating determining a topological second image according to some embodiments of the present disclosure. In some embodiments, the flow 800 can be performed by a processing device. For example, the flow 800 can be stored in a storage device (such as a self-provided storage unit of the processing device or an external storage device) in the form of a program or instructions, which when executed, can implement the flow 800. As shown in Figure 8 the flow 800 can include the following operations.

[0163] At step 802, sample material density data is determined based on the sample base material density image.

[0164] At step 804, sample topological data difference is determined based on the sample first topological data and the sample second topological data.

[0165] At step 806, material decomposition matrix difference is determined based on the sample material density data and the sample topological data difference.

[0166] At step 808, image difference is determined based on the material decomposition matrix difference and the sample base material density image.

[0167] At step 810, the topological second image is determined based on the sample first image and the image difference.

[0168] It can be understood that, since the topological second image is similar to the second image, both of which can be calculated by the first image and the topological data, the difference between determining the topological second image and determining the second image lies in the data used in the calculation, but the calculation process can be mutually referenced. Specifically, the second image is calculated based on the first image, the base material density image and the topological data, while the topological second image is calculated based on the sample first image, the sample base material density image and the sample topological data. Therefore, as to the details of the process of determining the topological second image, reference can be made to the related description of the process of determining the second image. Figure 8 Figure 2 and Figure 3 The related description of the process of determining the second image is not repeated here.

[0169] It should be noted that the above description of the processes is only for example and illustration, and does not limit the scope of the present specification. Those skilled in the art can make various modifications and changes to the processes under the guidance of the present specification. However, these modifications and changes are still within the scope of the present specification. For example, changes can be made to the steps of the processes described in the present specification, such as adding preprocessing steps and storage steps.

[0170] Figure 9 is an exemplary block diagram of an image imaging system according to some embodiments of the present specification. As shown in Figure 9 the system can include an acquisition module 910, a first determination module 920, a second determination module 930, a third determination module 940 and a training module 950.

[0171] The acquisition module 910 can be used to acquire the first image and the topological data of the target object.

[0172] ​The target object refers to an object that needs to be processed by the image processing. The target object can include a patient or other medical experimental object (e.g., a test mouse or other animals) and the like. The target object can also be a part of a patient or other medical experimental object, including organs and / or tissues, such as a heart, a lung, a rib, an abdominal cavity, and the like.

[0173] In some embodiments, the topological data includes first topological data and second topological data, the first topological data corresponding to the second topological data, the first image and the first topological data being obtained under a first radiation dose scan, and the second topological data being obtained under a second radiation dose scan.

[0174] In some embodiments, the obtaining module 910 can obtain the first image and the topological data of the target object by reading from an imaging device, a database, a storage device, or calling a related data interface.

[0175] The first determining module 920 can be configured to determine a base material density image corresponding to the first image.

[0176] In some embodiments, the first determining module 920 can process the first image by using a preset image processing model to determine the base material density image corresponding to the first image.

[0177] In some embodiments, the first determining module 920 can input the first image into a preset image processing model to output the base material density image by the image processing model. In some embodiments, the image processing model can include a U-net model, a V-net model, and the like based on deep learning.

[0178] The second determining module 930 can be configured to determine a topological data difference based on the first topological data and the second topological data.

[0179] The topological data difference refers to a difference between energy topological data of two different energy levels. In some embodiments, the second determining module 930 can subtract the first topological data from the second topological data to obtain the topological data difference.

[0180] The third determining module 940 can be configured to determine a second image corresponding to the first image based on the first image, the base material density image, and the topological data difference.

[0181] In some embodiments, the third determining module 940 can determine an image decomposition matrix difference based on the base material density image and the topological data difference, determine an image difference based on the base material density image and the image decomposition matrix difference, and determine the second image based on the first image and the image difference.

[0182] In some embodiments, the third determining module 940 can determine material density data based on the base material density image; determine the image decomposition matrix difference based on the material density data and the topological data difference.

[0183] In some embodiments, the image imaging system can further include a training module 950. The training module 950 can be configured to train the image processing model.

[0184] The training module 950 can be configured to obtain a plurality of training samples, wherein each of the plurality of training samples includes a sample first image.

[0185] The training module 950 can be configured to input the training sample into the image processing model, and determine a sample base material density image corresponding to the sample first image.

[0186] The training module 950 can be configured to adjust parameters of the image processing model based on one or more of the label base material density image, the sample first image, a sample second image, sample topological data, and the sample base material density image, and optimize a target loss function as a training target, to obtain a trained image processing model; wherein the sample second image corresponds to the sample first image.

[0187] It should be noted that the training module 950 is optional in the image processing system 900, and in some embodiments, the image processing system 900 can also not include the training module 950.

[0188] For specific descriptions of the modules of the image imaging system, reference can be made to the related descriptions in the flowchart part of the present specification, for example, Figures 2 to 8 .

[0189] It should be understood that Figure 9The illustrated system and its modules can be implemented in various ways. For instance, in some embodiments, the system and its modules can be implemented in hardware, software, or a combination of software and hardware. The hardware components can be implemented with special logic, while the software components can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art can understand that the above-described methods and systems can be implemented using computer-executable instructions and / or included in processor control codes, such as provided on a carrier medium, such as a disk, CD or DVD-ROM, programmable memory (firmware), or data carrier such as an optical or electrical signal carrier. The system and its modules of the present specification can be implemented not only in hardware circuitry, such as very large scale integrated circuits or gate arrays, semiconductors, such as logic chips, transistors, or programmable hardware devices, such as field programmable gate arrays, programmable logic devices, etc., but also in software, for example, executed by various types of processors, and also by a combination of the above-mentioned hardware circuitry and software (e.g., firmware).

[0190] It should be noted that the above description of the training system for image processing model and its modules is for convenience of description only, and does not limit the present specification to the scope of the embodiments. It can be understood that, for those skilled in the art, after understanding the principles of the system, various modules can be combined or connected to form subsystems without departing from the principles. For example, in some embodiments, the first determination module 920 and the second determination module 930 can be different modules in a system, or a module can implement the functions of two or more modules described above. For example, the modules can share a storage module, and each module can have its own storage module. Variations such as these are within the scope of the present specification.

[0191] The beneficial effects that the embodiments of the present specification can bring include but are not limited to: (1) combining deep learning technology, using a trained image processing model to obtain the corresponding second image based on the single-energy first image; (2) the hardware implementation of the embodiments of the present specification is simple, since the first image and the first topological data obtained by using a low-dose ray to scan the target object, and only using a relatively high-dose ray to scan the target object at an angle, the second topological data can be obtained, reducing the radiation dose to the target object; (3) by introducing topological data, more information can be obtained in the calculation process, thereby improving the dual-energy CT value, and improving the accuracy of the dual-energy image post-processing result (4) before the first image is input into the image processing model, the first image can not be processed such as denoising, simplifying the process of obtaining the base material density image; (5) the output result of the image processing model is the base material density image, compared with using a deep learning model to obtain an estimated second image from the first image, and then obtaining the base material density image based on the first image and the estimated second image, the problem of serious degradation of image signal-to-noise ratio caused by using matrix inversion on the first image and the second image can be avoided or reduced, and the time spent on iterative material decomposition is saved.

[0192] It should be noted that different embodiments can produce different beneficial effects, and in different embodiments, the beneficial effects that can be produced can be any one or a combination of the above, or any other beneficial effects that can be obtained.

[0193] The above has described the basic concepts, and it is obvious that the above detailed disclosure is only used as an example and does not limit the present specification. Although it is not explicitly stated here, those skilled in the art can make various modifications, improvements and corrections to the present specification. Such modifications, improvements and corrections are suggested in the present specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the present specification.

[0194] At the same time, specific words are used in the present specification to describe the embodiments of the present specification. As "one embodiment", "an embodiment", and / or "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of the present specification. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "one alternative embodiment" mentioned in different places in the present specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present specification can be properly combined.

[0195] Moreover, those skilled in the art will appreciate that the various aspects of the disclosure can be illustrated and described in connection with a number of various kinds of systems or circumstances, including any new and useful processes, machines, products, or compositions of matter, or any new and useful improvements thereof, as defined by the plain language of the claims. Accordingly, the various aspects of the disclosure can be implemented in whole or in part any number of hardware, software codes, firmware codes, and / or other programming codes. Such hardware, software codes, firmware codes, and / or other programming codes can be referred to as a "data block", "module", "engine", "unit", "component", or "system". In addition, aspects of the disclosure can be embodied as a computer program product on one or more computer readable media (moving or physical) having computer readable program code embodied therein.

[0196] Computer storage media can include a propagated data signal with the computer program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal can take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. Computer storage media can be any media that can be accessed by a computer. By way of example, and not limitation, such computer storage media can comprise RAM, ROM, EEPROM, CD-ROM or any other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired computer program code in the form of computer-readable program code means, and which can be accessed by a computer. Also, any connection is properly termed a computer storage medium. For example, if the computer program code is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, or twisted pair, as exemplary signal bearing media, "computer storage medium" and "computer program medium" can be used interchangeably.

[0197] The computer program code can also be loaded onto a computer, other programmable data processing apparatus, or computer-readable storage media to cause a series of operational steps to be performed on the computer, other programmable apparatus or other computer-readable storage media to produce a computer implemented process such that the code which implement the process is

[0198] Furthermore, the order of the processing elements and sequences described in this specification are not intended to be construed as a limitation, unless specifically stated, but are included to provide a complete description of one or more embodiments of the present specification. Regardless of the particular sequence of processing elements and sequences, however, the description herein of a process should be understood to include any and all combinations of one or more elements of a process independently selected from each sequence. For example, although the system components described above can be implemented by hardware devices, they can also be implemented by software solutions, such as installing the described system on an existing server or mobile device.

[0199] Similarly, it is to be noticed that the term "comprising", used in the description, is not intended to exclude other elements or steps. It is to be understood that the description and the examples are intended to be illustrative, but not limiting, of the scope of the present specification. Thus, the scope of the present specification should be given by the appended claims, along with their full scope of equivalents, and not by an restricting interpretation of the description or the examples.

[0200] Some embodiments use numerical designations to describe components, quantities of attributes. It is to be understood that such numerical designations used in the description of embodiments are, in some examples, modified by the adjectives "about", "approximately", or "generally". Unless otherwise stated, "about", "approximately", or "generally" indicates that the stated numerical value is allowed ±20% variation. Accordingly, numerical values used in the description and claims are approximations that can vary depending on the desired properties of the individual embodiments. In some embodiments, numerical values should be considered in the context of the number of significant figures used in the description and claims. Although the numerical ranges and parameters setting forth the broadest scope of the embodiments herein are approximations, the numerical values set forth in the specific examples are reported as precisely as practicable. The numerical values set forth in the specific examples are provided to give a general understanding of the embodiments.

[0201] Each patent, patent application, patent publication, and other material cited in this specification is incorporated herein by reference in its entirety. In the event of inconsistencies between the disclosure of this specification and the documents, articles, or other materials incorporated by reference, the disclosure of this specification will prevail. In the event of inconsistencies between the disclosure of this specification and the claims, the claims will prevail. It is specifically intended that the description, definitions, and / or terminology used in the materials incorporated by reference into this specification be interpreted and used in accordance with the description, definitions, and / or terminology used in the present specification. Nothing herein is to be construed as an admission that the present specification is not entitled to antedate such material by virtue of prior application.

[0202] Finally, it should be understood that the embodiments described herein are only given by way of example and that other modifications can occur to persons skilled in the art. Therefore, the scope of the present description is not intended to be limited to the embodiments described herein but is only limited by the claims that follow.

Claims

1. An image imaging method, characterized in that, The method includes: Acquire a first image and topological data of a target object, wherein the topological data includes first topological data and second topological data, the first topological data corresponds to the second topological data, the first image and the first topological data are obtained by the target object under a first radiation dose scan, and the second topological data are obtained by the target object under a second radiation dose scan; Determining the base material density image corresponding to the first image includes: inputting the first image into a pre-trained image processing model to determine the base material density image corresponding to the first image; Based on the first topology data and the second topology data, the topology data differences are determined; Based on the differences between the first image, the base material density image, and the topological data, a second image corresponding to the first image is determined; The image processing model is trained in the following way: Acquire multiple training samples, wherein each of the multiple training samples includes a first image of the sample; The training samples are input into the image processing model to determine the sample base material density image corresponding to the first image of the sample. A target loss function is constructed based at least on the label-based material density image and the sample-based material density image. The parameters of the image processing model are adjusted with the optimization of the target loss function as the training objective to obtain a trained image processing model. The target loss function term constructed based on the label-based material density image and the sample-based material density image is used to characterize the product of the square of the L2 norm of the difference between the label-based material density image and the sample-based material density image and the reciprocal of the total number of training samples used for model training.

2. The method according to claim 1, characterized in that, The step of determining the second image corresponding to the first image based on the differences between the first image, the base material density image, and the topological data includes: Based on the differences between the base material density image and the topological data, the differences in the image decomposition matrix are determined; The image difference is determined based on the difference between the base material density image and the image decomposition matrix; The second image is determined based on the differences between the first image and the image.

3. The method according to claim 2, characterized in that, The step of determining the image decomposition matrix difference based on the difference between the base material density image corresponding to the first image and the topological data includes: Based on the density image of the base material, determine the material density data; The differences in the image decomposition matrix are determined based on the differences between the material density data and the topology data.

4. The method according to claim 1, characterized in that, Obtaining the first topology data includes: Obtain the first topology data from the scan data corresponding to the first image; or The target object is scanned and acquired at the first radiation dose.

5. The method according to claim 1, characterized in that, The training of the image processing model also includes: The target loss function is constructed based on one or more of the sample first image, sample second image, and sample topology data; wherein the sample second image and the sample first image are derived from scanning the same target object.

6. The method according to claim 5, characterized in that, The target loss function includes: Base material density image loss function; And any one or a combination of the first image loss function, the second image loss function, and the third image loss function; Wherein, the base material density image loss function is determined at least based on the label base material density image, the first image loss function is determined at least based on the first sample image, the second image loss function is determined at least based on the second sample image, and the third image loss function is determined based on the second sample image and sample topology data; The sample topology data includes sample first topology data and sample second topology data, and the sample first topology data and the sample second topology data correspond to each other.

7. The method according to claim 1, characterized in that, The first image is a low-energy image, and the second image is a high-energy image; The low-energy image is obtained by scanning the target object at a first radiation dose, and the high-energy image is obtained by scanning the target object at a second radiation dose; the first radiation dose is lower than the second radiation dose.

8. An image imaging system, characterized in that, The system includes: The acquisition module is used to acquire a first image and topological data of a target object, wherein the topological data includes first topological data and second topological data, the first topological data corresponds to the second topological data, the first image and the first topological data are obtained by the target object under a first radiation dose scan, and the second topological data are obtained by the target object under a second radiation dose scan; The first determining module is used to determine the base material density image corresponding to the first image, including: inputting the first image into a pre-trained image processing model to determine the base material density image corresponding to the first image; The second determining module is used to determine the topology data differences based on the first topology data and the second topology data; The third determining module is used to determine the second image corresponding to the first image based on the first image, the base material density image, and the topological data difference; The image processing model is trained in the following way: Acquire multiple training samples, wherein each of the multiple training samples includes a first image of the sample; The training samples are input into the image processing model to determine the sample base material density image corresponding to the first image of the sample. A target loss function is constructed based at least on the label-based material density image and the sample-based material density image. The parameters of the image processing model are adjusted with the optimization of the target loss function as the training objective to obtain a trained image processing model. The target loss function term constructed based on the label-based material density image and the sample-based material density image is used to characterize the product of the square of the L2 norm of the difference between the label-based material density image and the sample-based material density image and the reciprocal of the total number of training samples used for model training.

9. An image imaging apparatus, comprising at least one storage medium and at least one processor, the at least one storage medium being configured to store computer instructions; the at least one processor being configured to execute the computer instructions to implement the method as described in any one of claims 1 to 7.

Images