Systems and methods for breast imaging
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2026-02-10
- Publication Date
- 2026-08-13
Smart Images

Figure CN2026078211_13082026_PF_FP_ABST
Abstract
Description
SYSTEMS AND METHODS FOR BREAST IMAGINGCROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to Chinese Patent Application No. 202510144575.9, filed on February 10, 2025, the contents of which are incorporated herein by reference.TECHNICAL FIELD
[0002] The present disclosure relates to the field of medical technology, and more particularly, relates to systems and methods for breast imaging.BACKGROUND
[0003] During the diagnosis of breast lesions, precise insertion of a puncture needle into the region of interest (ROI) is essential. This procedure involves guiding the needle, based on breast imaging, to collect a tissue sample from the ROI for biopsy. Therefore, there is a need for breast imaging systems and methods capable of producing images with high accuracy and clarity to support subsequent needle guidance and diagnostic procedures.SUMMARY
[0004] According to an aspect of the present disclosure, a method for breast imaging is provided. The method may be implemented on a computing device having at least one processor and at least one storage device. The method may include obtaining a plurality of projection image sets of a target breast collected using a multi-energy digital breast tomosynthesis (DBT) device. Each of the plurality of projection image sets may correspond to one of a plurality of energy levels, each projection image set may comprise a plurality of projection images that are collected at a plurality of projection angles. The method may include generating, based on the plurality of projection image sets, at least one three-dimensional (3D) subtraction image of the target breast. The method may further include generating a rendered 3D subtraction image by rendering the 3D subtraction image.
[0005] In some embodiments, the methods may further comprise identifying a region of interest (ROI) from the rendered 3D subtraction image; adding an annotation associated with the identified ROI on the rendered 3D subtraction image; and displaying, via a user interface, the rendered 3D subtraction image with the annotation.
[0006] In some embodiments, the ROI may comprise a lesion region, and the annotation associated with the identified ROI may indicate a puncture route of the lesion region.
[0007] In some embodiments, the methods may further comprise generating a diagnosis result of the target breast by analyzing the rendered 3D subtraction image.
[0008] In some embodiments, the rendered 3D subtraction image may include a 3D target breast model that is rotatable to display the target breast from different perspectives.
[0009] In some embodiments, the generating a rendered 3D subtraction image by rendering the 3D subtraction image may comprise preprocessing the 3D subtraction image to generate a preprocessed 3D subtraction image; and generating the rendered 3D subtraction image by performing pseudo-color rendering on an ROI in the preprocessed 3D subtraction image.
[0010] In some embodiments, each of the plurality of energy levels may be within a range of 20 keV to 50 keV.
[0011] In some embodiments, the plurality of projection angles may be within a limited angular range, the span of the limited angular range may not be greater than 50 degrees.
[0012] In some embodiments, the plurality of energy levels may comprise a first energy level and a second energy level, the projection image sets may comprise a first projection image set corresponding to the first energy level and a second projection image set corresponding to the second energy level, and the generating at least one 3D subtraction image of the target breast may comprise for each of the plurality of projection angles, generating a two-dimensional (2D) subtraction image by performing dual-energy subtraction on a first projection image corresponding to the projection angle in the first projection image set and a second projection image corresponding to the projection angle in the second projection image set; generating a 3D subtraction image by reconstructing the plurality of 2D subtraction images corresponding to the plurality of projection angles.
[0013] In some embodiments, the artifact correction may be performed during the generation process of the 3D subtraction image via a constraint term configured to constrain a signal distribution of an ROI along a depth direction.
[0014] In some embodiments, the plurality of 2D subtraction images corresponding to the plurality of projection angles may be reconstructed through an iterative reconstruction process, and a current iteration in the iterative reconstruction process may comprise determining the signal distribution of the ROI along the depth direction in a current image to be updated in the current iteration; determining the constraint term based on the signal distribution; generating an updated image by updating the current image based on the plurality of 2D subtraction images and the constraint term; and designating the updated image as a current image of a next iteration, or designating the updated image as the 3D subtraction image.
[0015] In some embodiments, the updating the current image may be performed based on a reconstruction target comprising a data fidelity term and the constraint term, the data fidelity term may be associated with a difference between forward projection of the current image and the plurality of 2D subtraction images.
[0016] In some embodiments, the constraint term may be a directional total variation.
[0017] In some embodiments, the generating a 3D subtraction image by reconstructing the plurality of 2D subtraction images corresponding to the plurality of projection angles may comprise for each of the plurality of 2D subtraction images, segmenting an ROI from the 2D subtraction image to generate a local image of the ROI; and reconstructing the plurality of local images of the ROI to generate the 3D subtraction image.
[0018] In some embodiments, the plurality of energy levels may comprise a first energy level and a second energy level, the projection image sets may comprise a first projection image set corresponding to the first energy level and a second projection image set corresponding to the second energy level, and the generating at least one 3D subtraction image of the target breast may comprise generating a first 3D reconstruction image corresponding to the first energy level by reconstructing first projection images in the first projection image set; generating a second 3D reconstruction image corresponding to the second energy level by reconstructing second projection images in the second projection image set; and generating the 3D subtraction image by performing dual-energy subtraction on the first 3D reconstruction image and the second 3D reconstruction image.
[0019] In some embodiments, the multi-energy DBT device may comprise a radiation source that is capable of switching between the plurality of energy levels, or the multi-energy DBT device may comprise a plurality of radiation sources respectively corresponding to the plurality of energy levels.
[0020] In some embodiments, the multi-energy DBT device may comprise an energy-resolving detector that is capable of distinguishing radiation at the plurality of energy levels.
[0021] In some embodiments, the energy-resolving detector may be a photon-counting flat panel detector, the plurality of energy levels comprise at least three energy levels, and the generating, based on the plurality of projection image sets, at least one 3D subtraction image of the target breast may comprise determining a characteristic energy level of an ROI of the target breast; determining, based on the characteristic energy levels, two target energy levels from the at least three energy levels; and generating a target 3D subtraction image for the ROI based on the projection image sets corresponding to the two target energy levels.
[0022] In some embodiments, the energy-resolving detector may be a photon-counting flat panel detector, the plurality of energy levels may comprise at least three energy levels, and the generating, based on the plurality of projection image sets, at least one 3D subtraction image of the target breast may comprise determining a plurality of energy level pairs among the at least three energy levels; and for each energy level pair, generating a 3D subtraction image based on the projection image sets corresponding to the energy level pairs.
[0023] According to another aspect of the present disclosure, a system for breast imaging is provided. The system may include at least one storage device including a set of instructions, and at least one processor configured to communicate with the at least one storage device. When executing the set of instructions, the at least one processor may be configured to direct the system to perform operations. The operations may include obtaining a plurality of projection image sets of a target breast collected using a multi-energy digital breast tomosynthesis (DBT) device. Each of the plurality of projection image sets may correspond to one of a plurality of energy levels, each projection image set may comprise a plurality of projection images that are collected at a plurality of projection angles. The operations may include generating, based on the plurality of projection image sets, at least one three-dimensional (3D) subtraction image of the target breast. The operations may further include generating a rendered 3D subtraction image by rendering the 3D subtraction image.
[0024] According to still another aspect of the present disclosure, a non-transitory computer readable medium is provided. The non-transitory computer readable medium may include executable instructions. When executed by at least one processor, the at least one processor may be directed to perform a method. The method may include obtaining a plurality of projection image sets of a target breast collected using a multi-energy digital breast tomosynthesis (DBT) device. Each of the plurality of projection image sets may correspond to one of a plurality of energy levels, each projection image set may comprise a plurality of projection images that are collected at a plurality of projection angles. The method may include generating, based on the plurality of projection image sets, at least one three-dimensional (3D) subtraction image of the target breast. The method may further include generating a rendered 3D subtraction image by rendering the 3D subtraction image.
[0025] Additional features will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following and the accompanying drawings or may be learned by production or operation of the examples. The features of the present disclosure may be realized and attained by practice or use of various aspects of the methodologies, instrumentalities and combinations set forth in the detailed examples discussed below.BRIEF DESCRIPTION OF THE DRAWINGS
[0026] The present disclosure is further illustrated in terms of exemplary embodiments. These exemplary embodiments are described in detail according to the drawings. These embodiments are non-limiting exemplary embodiments, in which like reference numerals represent similar structures, wherein:
[0027] FIG. 1 is a schematic diagram illustrating an exemplary imaging system according to some embodiments of the present disclosure;
[0028] FIG. 2 is a block diagram illustrating an exemplary processing device according to some embodiments of the present disclosure;
[0029] FIG. 3 is a flowchart illustrating an exemplary process for breast imaging according to some embodiments of the present disclosure;
[0030] FIG. 4 is a schematic diagram illustrating an exemplary process for breast imaging according to some embodiments of the present disclosure;
[0031] FIG. 5 is a schematic diagram illustrating an exemplary 3D subtraction image according to some embodiments of the present disclosure;
[0032] FIG. 6A is a schematic diagram illustrating an exemplary non-focused layer of a subject during image reconstruction according to some embodiments of the present disclosure;
[0033] FIG. 6B is a schematic diagram illustrating an exemplary focused layer of a subject during image reconstruction according to some embodiments of the present disclosure;
[0034] FIG. 7 is a schematic diagram illustrating an exemplary rendered 3D subtraction image seen from a certain perspective according to some embodiments of the present disclosure;
[0035] FIG. 8 is a schematic diagram illustrating an exemplary process for generating a 3D subtraction image according to some embodiments of the present disclosure;
[0036] FIG. 9 is a schematic diagram illustrating another exemplary process for generating a 3D subtraction image according to some embodiments of the present disclosure;
[0037] FIG. 10 is a flowchart illustrating an exemplary process for generating a target 3D subtraction image according to some embodiments of the present disclosure; and
[0038] FIG. 11 is a flowchart illustrating an exemplary process for generating a 3D subtraction image according to some embodiments of the present disclosure.DETAILED DESCRIPTION
[0039] To more clearly illustrate the technical solutions related to the embodiments of the present disclosure, a brief introduction of the drawings referred to the description of the embodiments is provided below. Obviously, the drawings described below are only some examples or embodiments of the present disclosure. Those having ordinary skills in the art, without further creative efforts, may apply the present disclosure to other similar scenarios according to these drawings. Unless obviously obtained from the context or the context illustrates otherwise, the same numeral in the drawings refers to the same structure or operation.
[0040] It should be understood that “system, ” “device, ” “unit, ” and / or “module” as used herein is a manner used to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other words serve the same purpose, the words may be replaced by other expressions.
[0041] As shown in the present disclosure and claims, the words “one, ” “a, ” “a kind, ” and / or “the” are not especially singular but may include the plural unless the context expressly suggests otherwise. In general, the terms “comprise, ” “comprises, ” “comprising, ” “include, ” “includes, ” and / or “including, ” merely prompt to include operations and elements that have been clearly identified, and these operations and elements do not constitute an exclusive listing. The methods or devices may also include other operations or elements.
[0042] It will be understood that, although the terms “first, ” “second, ” “third, ” “fourth, ” etc., may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments of the present invention.
[0043] Spatial and functional relationships between elements are described using various terms, including “connected, ” “engaged, ” “interfaced, ” and “coupled. ” Unless explicitly described as being “direct, ” when a relationship between first and second elements is described in the present disclosure, that relationship includes a direct relationship where no other intervening elements are present between the first and second elements, and also an indirect relationship where one or more intervening elements are present (either spatially or functionally) between the first and second elements. In contrast, when an element is referred to as being “directly” connected, engaged, interfaced, or coupled to another element, there are no intervening elements present. Other words used to describe the relationship between elements should be interpreted in a like fashion (e.g., “between, ” versus “directly between, ” “adjacent, ” versus “directly adjacent, ” etc. ) . As used herein, the term “and / or” includes any and all combinations of one or more of the associated listed items.
[0044] The flowcharts used in the present disclosure illustrate operations that systems implement according to some embodiments of the present disclosure. It should be understood that the previous or subsequent operations may not be accurately implemented in order. Instead, each step may be processed in reverse order or simultaneously. Meanwhile, other operations may also be added to these processes, or a certain step or several steps may be removed from these processes.
[0045] In the present disclosure, the term “image” may refer to a two-dimensional (2D) image, a three-dimensional (3D) image, or a four-dimensional (4D) image (e.g., a time series of 3D images) . In some embodiments, the term “image” may refer to an image of a region (e.g., a region of interest (ROI) ) of a subject. In some embodiments, the image may be a medical image, an optical image, etc.
[0046] In the present disclosure, a representation of a subject (e.g., an object, a patient, or a portion thereof) in an image may be referred to as “subject” for brevity. For instance, a representation of an organ, tissue (e.g., a heart, a liver, a lung) , or an ROI in an image may be referred to as the organ, tissue, or ROI, for brevity. Further, an image including a representation of a subject, or a portion thereof, may be referred to as an image of the subject, or a portion thereof, or an image including the subject, or a portion thereof, for brevity. Still further, an operation performed on a representation of a subject, or a portion thereof, in an image may be referred to as an operation performed on the subject, or a portion thereof, for brevity. For instance, a segmentation of a portion of an image including a representation of an ROI from the image may be referred to as a segmentation of the ROI for brevity.
[0047] For illustration purposes, the following description is provided to help better understand an imaging process. It is understood that this is not intended to limit the scope of the present disclosure. For persons having ordinary skills in the art, a certain amount of variations, changes and / or modifications may be deducted under guidance of the present disclosure. Those variations, changes and / or modifications do not depart from the scope of the present disclosure.
[0048] In breast diagnosis, it is necessary to puncture the breast and collect living tissue from a lesion region for further pathological analysis. High-precision breast imaging is critical for accurately guiding such puncture procedures. Digital Breast Tomosynthesis (DBT) is a widely adopted breast imaging technique that generates a three-dimensional volume of the breast. This is achieved by acquiring a series of low-dose, two-dimensional projection images over a limited angular range, which are then reconstructed into thin slices. While DBT mitigates the issue of tissue superposition inherent in 2D mammography, its reliance on single-energy imaging often results in low contrast, particularly in dense breast tissue. This limitation can obscure subtle lesions and compromise detection capability.
[0049] Another advanced technology, contrast-enhanced spectral mammography (CESM) , employs dual-energy imaging to improve lesion conspicuity. After intravenous injection of an iodine-based contrast agent, low-energy and high-energy images are acquired from a single angle and subsequently processed to create a subtracted “iodine map” highlighting vascularized lesions. Although CESM enhances lesion detection, the resulting subtraction images are fundamentally two-dimensional and lack the spatial depth information crucial for precise lesion localization and characterization.
[0050] Consequently, a significant technological gap exists. Current solutions fail to provide both the three-dimensional anatomical context of DBT and the functional, contrast-enhanced information of CESM within a single framework. There is a clear need for a breast imaging system that simultaneously enables 3D reconstruction and offers enriched, accurate lesion characterization to improve diagnostic confidence, lesion detection rates, and the accuracy of subsequent interventional procedures such as needle guidance.
[0051] To solve the above problems, the present disclosure provides systems and methods that lie in the integration of DBT and CESM technologies, proposing a novel multi-energy DBT imaging technique to synergistically combine their respective advantages. Specifically, the methods may include obtaining a plurality of projection image sets of a target breast collected using a multi-energy DBT device. Each of the plurality of projection image sets may correspond to one of a plurality of different energy levels, and each projection image set may comprise a plurality of projection images that are collected at a plurality of projection angles within a limited angular range at the corresponding energy level. The methods may include generating, based on the plurality of projection image sets, at least one three-dimensional (3D) subtraction image of the target breast. In some embodiments, the methods may further include generating a rendered 3D subtraction image by rendering the 3D subtraction image.
[0052] The proposed methods retain the fundamental DBT principle of “acquiring multi-angle projections over a limited angular range followed by 3D image reconstruction. ” This allows the methods to provide detailed stereoscopic spatial information of lesions, overcoming the lack of depth information in conventional CESM. Concurrently, the proposed methods incorporate the core CESM methodology of "performing dual-energy exposure and subsequent image subtraction. " This integration enables the visualization of functional information by highlighting vascularized lesions that uptake contrast agent, thereby addressing the low-contrast issue inherent in single-energy DBT, especially in dense breast tissue. Ultimately, this integrated approach leads to a substantial improvement in diagnostic accuracy and offers a more reliable foundation for precise needle guidance during interventions.
[0053] FIG. 1 is a schematic diagram illustrating an exemplary imaging system 100 according to some embodiments of the present disclosure.
[0054] As shown in FIG. 1, in some embodiments, the imaging system 100 (e.g., a breast imaging system) may include an imaging device 110, a network 120, one or more terminal devices 130, a processing device 140, and a storage device 150. In some embodiments, the components of the imaging system 100 may be connected to each other via the network 120. Alternatively or additionally, the components of the imaging system 100 may be directly connected to each other. Merely by way of example, the imaging device 110 may be connected to the processing device 140 through the network 120, as illustrated in FIG. 1. As another example, the imaging device 110 may be connected to the processing device 140 directly.
[0055] The imaging device 110 may be configured to generate or provide image data by scanning a subject or at least a part of the subject. For example, the imaging device 110 may scan the subject or a portion thereof that is located within its detection region and generate the image data (e.g., projection images) relating to the subject or the portion thereof. In some embodiments, the subject may include a biological object and / or a non-biological object. The biological object may be a human being, an animal, or a specific portion, organ, and / or tissue thereof. For example, the subject may be a target breast. In some embodiments, the subject may be a man-made composition of organic and / or inorganic matters that are with or without life. The terms “object” and “subject” are used interchangeably in the present disclosure.
[0056] In some embodiments, the imaging device 110 may be a DBT device. The DBT device may be configured to collect the projection images of the target breast using a DBT technique. In some embodiments, the DBT device may include a breast compression assembly, a radiation emitting assembly 113, and a detector assembly 114. The breast compression assembly may be configured to support and compress the target breast. As shown in FIG. 1, the breast compression assembly may include a support platform 111 and a compression plate 112. The support platform 111 may be configured to place the target breast (e.g., the left breast or the right breast) . The compression plate 112 may be disposed opposite to the support platform 111, and may be configured to compress (squeeze) the target breast, thereby causing the target breast to deform.
[0057] The radiation emitting assembly 113 may emit a plurality of radiation beams (e.g., X-rays) to the subject. The detector assembly 114 may collect scan data by detecting the radiation beams passing through a detection region of the detector assembly 114. For example, the detector assembly 114 may convert radiation photons of the radiation beams into an electronic signal. The electronic signal may be transmitted to a computing device (e.g., the processing device 140) for processing, or transmitted to a storage device (e.g., the storage device 150) for storage.
[0058] During a DBT scan, the target breast is placed on the support platform 111, and the compression plate 112 moves down to compress the target breast. The radiation emitting assembly 113 (e.g., the radiation source) rotates, and the detector assembly 114 is mounted on or within the support platform 111 and remains stationary. For example, the detector assembly 114 is fixed, and the radiation emitting assembly 113 rotates within a limited angular range at a rotation step to irradiate the target breast from different projection angles.
[0059] Accordingly, the detector assembly 114 may collect two-dimensional (2D) projection images at the plurality of projection angles within the limited angular range. A three-dimensional (3D) image including thin slices may be generated by reconstructing the 2D projection images based on an image reconstruction technique (or algorithm) . The imaging system 100 including the DBT device may also be referred to as a DBT system. Through the DBT technique, the problem of tissue overlap can be solved, thereby enhancing imaging quality and improving a detection rate of lesions.
[0060] As another example, the imaging device 110 may be a multi-energy DBT device. The multi-energy DBT device may be the DBT device that can perform a DBT scan at a plurality of different energy levels. In some embodiments, the imaging device 110 may be configured to collect a plurality of projection image sets of the target breast. Each of the plurality of projection image sets may correspond to one of the plurality of different energy levels, and each projection image set may comprise a plurality of projection images that are collected at the plurality of projection angles within the limited angular range at the corresponding energy level.
[0061] In some embodiments, the multi-energy DBT device is configured by modifying the DBT device with a radiation emitting assembly 113 and / or a detector assembly 114 adapted for multi-energy imaging. Specifically, the radiation emitting assembly 113 may comprise either a radiation source operable to switch between a plurality of energy levels, or a plurality of radiation sources each corresponding to one of the energy levels. This configuration is thereby capable of emitting a plurality of radiation beams at the respective energy levels.
[0062] As another example, the detector assembly 114 may collect scan data (e.g., the plurality of projection image sets) of the plurality of energy levels by detecting the radiation beams passing through the detection region. For instance, the detector assembly 114 may include an energy-resolving detector that is capable of distinguishing radiation at the plurality of energy levels. Exemplary energy-resolving detectors may include a multi-layer flat panel detector or a photon-counting flat panel detector.
[0063] The multi-layer flat panel detector may include a plurality of layers, and each of the plurality of energy levels may correspond to one of the plurality of layers. For example, the plurality of energy levels may comprise a first energy level and a second energy level, and the multi-layer flat panel detector may include a first layer and a second layer. The first layer of the multi-layer flat panel detector may be configured to collect radiation photons of the radiation beams corresponding to the first energy level, and the second layer of the multi-layer flat panel detector may be configured to collect radiation photons of the radiation beams corresponding to the second energy level. In some embodiments, a filter layer may be disposed between the first layer and the second layer. A material of the filter layer may include copper or tin. For example, when the radiation emitting assembly 113 emits the radiation beams with a relatively high energy level, the first layer of the multi-layer flat panel detector may detect the radiation beams of the first energy level passing through the target breast, and the second layer of the multi-layer flat panel detector may detect the radiation beams of the second energy level passing through the target breast and the filter layer.
[0064] By using the multi-layer flat panel detector, the projection image sets corresponding to the different energy levels can be simultaneously collected in a single exposure, thereby avoiding misalignment artifacts caused by two scans. In this way, the problem of image position mismatch caused by consecutive acquisitions by the radiation emitting assembly 113 can be solved.
[0065] The photon-counting flat panel detector may be configured to detect the energy of the radiation beams passing through the detection region. In some embodiments, the photon-counting flat panel detector may adopt a semiconductor conversion technology to directly convert the radiation photons (e.g., X-ray photons) of the radiation beams into electrical signals.
[0066] In some embodiments, before a multi-energy DBT scan, a contrast agent (e.g., iodine contrast agent, galactography contrast agent, etc. ) is injected through the upper limb vein, and then the target breast is compressed and exposed by the radiation beams corresponding to the plurality of different energy levels. Accordingly, the plurality of projection image sets corresponding to the plurality of different energy levels are collected at the plurality of projection angles within the limited angular range, and a 3D subtraction image is generated by performing a subtraction operation on the plurality of projection image sets. The imaging system 100 including the multi-energy DBT device may also be referred to as a multi-energy DBT system.
[0067] Through the subtraction operation, normal fat and glandular background of the target breast can be removed, and the 3D subtraction image can present hypervascular lesions (i.e., regions that take up the contrast agent) , thereby clearly displaying blood flow-enhanced lesions and improving the detection rate of lesions.
[0068] The network 120 may include any suitable network that can facilitate the exchange of information and / or data for the imaging system 100. In some embodiments, one or more components (e.g., the imaging device 110, the terminal (s) 130, the processing device 140, the storage device 150, etc. ) of the imaging system 100 may communicate information and / or data with one or more other components of the imaging system 100 via the network 120. For example, the processing device 140 may obtain the image data (e.g., the plurality of projection image sets) from the imaging device 110 via the network 120. As another example, the processing device 140 may obtain user instructions from the terminal (s) 130 via the network 120. In some embodiments, the network 120 may include one or more network access points. For example, the network 120 may 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 imaging system 100 may be connected to the network 120 to exchange data and / or information.
[0069] The terminal (s) 130 may include a mobile device 131, a tablet computer 132, a laptop computer 133, or the like, or any combination thereof. In some embodiments, the mobile device 131 may 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 may include a smart lighting device, a control device of an intelligent electrical apparatus, a smart monitoring device, a smart television, a smart video camera, an interphone, or the like, or any combination thereof. In some embodiments, the wearable device may include a bracelet, a footgear, eyeglasses, a helmet, a watch, clothing, a backpack, a smart accessory, or the like, or any combination thereof. In some embodiments, the mobile device may include a mobile phone, a personal digital assistant (PDA) , a gaming device, a navigation device, a point of sale (POS) device, a laptop, a tablet computer, a desktop, or the like, or any combination thereof. In some embodiments, the virtual reality device and / or the augmented reality device may include a virtual reality helmet, virtual reality glasses, a virtual reality patch, an augmented reality helmet, augmented reality glasses, an augmented reality patch, or the like, or any combination thereof. In some embodiments, the terminal (s) 130 may be part of the processing device 140.
[0070] The processing device 140 may process data and / or information obtained from one or more components (the imaging device 110, the terminal (s) 130, and / or the storage device 150) of the imaging system 100. For example, the processing device 140 may obtain the plurality of projection image sets of the target breast collected using the imaging device 110. As another example, the processing device 140 may generate at least one three-dimensional (3D) subtraction image of the target breast based on the plurality of projection image sets. As still another example, the processing device 140 may generate a rendered 3D subtraction image by rendering the 3D subtraction image. As yet another example, the processing device 140 may generate a diagnosis result of the target breast by analyzing the rendered 3D subtraction image.
[0071] In some embodiments, the processing device 140 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processing device 140 may be local or remote. For example, the processing device 140 may access information and / or data stored in the imaging device 110, the terminal (s) 130, and / or the storage device 150 via the network 120. As another example, the processing device 140 may be directly connected to the imaging device 110, the terminal (s) 130, and / or the storage device 150 to access stored information and / or data. In some embodiments, the processing device 140 may be implemented on a cloud platform. Merely by way of example, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-cloud, a multi-cloud, or the like, or any combination thereof.
[0072] In some embodiments, the processing device 140 may be implemented by a computing device. For example, the computing device may include a processor, a storage device, an input / output (I / O) , and a communication port. The processor may execute computer instructions (e.g., program codes) and perform functions of the processing device 140 in accordance with the techniques described herein. The computer instructions may include, for example, routines, programs, objects, components, data structures, procedures, modules, and functions, which perform particular functions described herein. In some embodiments, the processing device 140, or a portion of the processing device 140 may be implemented by a portion of the terminal (s) 130.
[0073] The storage device 150 may store data / information obtained from the imaging device 110, the terminal (s) 130, and / or any other component of the imaging system 100. In some embodiments, the storage device 150 may include a mass storage, a removable storage, a volatile read-and-write memory, a read-only memory (ROM) , or the like, or any combination thereof. For example, the mass storage may include a magnetic disk, an optical disk, a solid-state drive, etc. The removable storage may include a flash drive, a floppy disk, an optical disk, a memory card, a zip disk, a magnetic tape, etc. In some embodiments, the storage device 150 may store one or more programs and / or instructions to perform exemplary methods described in the present disclosure.
[0074] In some embodiments, the storage device 150 may be connected to the network 120 to communicate with one or more other components in the imaging system 100 (e.g., the processing device 140, the terminal (s) 130, etc. ) . One or more components in the imaging system 100 may access the data or instructions stored in the storage device 150 via the network 120. In some embodiments, the storage device 150 may be directly connected to or communicate with one or more other components in the imaging system 100 (e.g., the processing device 140, the terminal (s) 130, etc. ) . In some embodiments, the storage device 150 may be part of the processing device 140.
[0075] It should be noted that the above description is merely provided for the purposes of illustration, and is not intended to limit the scope of the present disclosure. For persons having ordinary skills in the art, multiple variations and modifications may be made under the teachings of the present disclosure. Features, structures, methods, and other characteristics of the exemplary embodiments described herein may be combined in various ways to obtain additional and / or alternative exemplary embodiments. However, those variations and modifications do not depart from the scope of the present disclosure.
[0076] FIG. 2 is a block diagram illustrating an exemplary processing device 140 according to some embodiments of the present disclosure. In some embodiments, the processing device 140 may be in communication with a computer-readable storage medium (e.g., the storage device 150 illustrated in FIG. 1) and execute instructions stored in the computer-readable storage medium. The processing device 140 may include an obtaining module 210 and a generation module 220.
[0077] The obtaining module 210 may be configured to obtain a plurality of projection image sets of a target breast collected using a multi-energy digital breast tomosynthesis (DBT) device. Each of the plurality of projection image sets may correspond to one of a plurality of different energy levels, each projection image set may comprise a plurality of projection images that are collected at a plurality of projection angles within a limited angular range at the corresponding energy level. More descriptions regarding the obtaining the plurality of projection image sets may be found elsewhere in the present disclosure. See, e.g., operation 302 and relevant descriptions thereof.
[0078] The generation module 220 may be configured to generate, based on the plurality of projection image sets, at least one three-dimensional (3D) subtraction image of the target breast. More descriptions regarding the generation of the at least one 3D subtraction image may be found elsewhere in the present disclosure. See, e.g., operation 304 and relevant descriptions thereof.
[0079] In some embodiments, the generation module 220 may be further configured to generate a rendered 3D subtraction image by rendering the 3D subtraction image. More descriptions regarding the generation of the rendered 3D subtraction image may be found elsewhere in the present disclosure. See, e.g., operation 306 and relevant descriptions thereof.
[0080] It should be noted that the above description regarding the processing device 140 is merely provided for the purposes of illustration, and is not intended to limit the scope of the present disclosure. For persons having ordinary skills in the art, multiple variations or modifications may be made under the teachings of the present disclosure. However, those variations and modifications do not depart from the scope of the present disclosure. For example, the processing device 140 may include a storage module configured to store data generated by the above-mentioned modules of the processing device 140. As another example, one or more modules may be integrated into a single module to perform the functions thereof.
[0081] FIG. 3 is a flowchart illustrating an exemplary process 300 for breast imaging according to some embodiments of the present disclosure. Process 300 may be implemented in the imaging system 100 illustrated in FIG. 1. For example, the process 300 may be stored in the storage device 150 in the form of instructions (e.g., an application) , and invoked and / or executed by the processing device 140. As shown in FIG. 3, the process 300 includes operations 302 and 304. In some embodiments, the process 300 further includes operation 306. In some embodiments, the process 300 further includes operation 308.
[0082] In 302, the processing device 140 (e.g., the obtaining module 210) may obtain a plurality of projection image sets of a target breast collected using a multi-energy DBT device.
[0083] The target breast refers to a breast to be imaged / diagnosed.
[0084] The multi-energy DBT device refers to a breast imaging system that combines a DBT technique with a multi-energy technique. More descriptions regarding the multi-energy DBT device may be found elsewhere in the present disclosure. See, e.g., FIG. 1 and relevant descriptions thereof.
[0085] In some embodiments, each of the plurality of projection image sets may correspond to one of the plurality of different energy levels. For example, the plurality of energy levels may include two or more energy levels. In some embodiments, each of the plurality of energy levels may be within a range of 20 keV to 50 keV. Merely by way of example, the plurality of energy levels may include a first energy level and a second energy level. The first energy level may correspond to low-energy radiation beams (e.g., X-rays) within a range of 26 keV to 33 keV, and the low-energy radiation beams may have a strong contrast and be configured to obtain images of soft tissues (e.g., fat tissues, etc. ) of the target breast. The second energy level may correspond to high-energy radiation beams within a range of 45 keV to 49 keV, and the high-energy radiation beams may have a strong penetrating power and be configured to obtain images of denser tissues (e.g., tumor tissues, etc. ) in the target breast.
[0086] In some embodiments, the processing device 140 may determine each of the plurality of energy levels according to medical requirements (e.g., scanning requirements, diagnostic requirements, etc. ) . For example, the processing device 140 may determine the first energy level within the range of 26 keV to 33 keV according to a requirement for determining the soft tissues of the target breast, and determine the second energy level within the range of 45 keV to 49 keV according to a requirement for determining the denser tissues in the target breast.
[0087] By designing a dedicated X-ray tube for breast imaging, each of the plurality of energy levels can be determined within the range of 20 keV to 50 keV. This range can be smaller than energy levels (e.g., 80 keV and 140keV) of a dual-energy CT system, and better matches the attenuation characteristics of soft tissues than the dual-energy CT system, thereby improving the soft tissue contrast of the target breast to meet the medical requirements of the breast imaging.
[0088] It should be noted that the above description regarding the high and low radiation beams is merely provided for the purposes of illustration, and not intended to limit the scope of the present disclosure. For example, three or more energy levels of radiation beams may be used to scan the target breast.
[0089] In some embodiments, each projection image set may comprise a plurality of projection images that are collected at a plurality of projection angles within a limited angular range at the corresponding energy level. A projection image refers to a 2D image generated by irradiating the target breast through radiation beams (e.g., X-rays) from a specific projection angle.
[0090] The limited angular range refers to a range or a span of a rotation angle of a radiation emitting assembly (e.g., the radiation emitting assembly 113, an X-ray tube of the radiation emitting assembly 113) during the collection of the plurality of projection image sets, and the rotation angle refers to an angle of the radiation emitting assembly rotating with respect to a reference direction (e.g., a straight line perpendicular to a detector plane of the detector assembly 114) . For example, the limited angular range may be a range or a span of the rotation angle where radiation beams emitted by the radiation emitting assembly are efficiently detected by the detector assembly.
[0091] In some embodiments, the span of the limited angular range may not be greater than a preset threshold. For example, the span of the limited angular range may be 10 degrees, 20 degrees, 30 degrees, 40 degrees, 50 degrees, etc. For instance, the span of the limited angular range may be a range of 0 degrees to 50 degrees. As another example, the span of the limited angular range may be a range of 10 degrees to 50 degrees. As still another example, the span of the limited angular range may be a range of 15 degrees to 50 degrees. As yet another example, the span of the limited angular range may be a range of 20 degrees to 40 degrees. As yet another example, the span of the limited angular range may be a range of 25 degrees to 35 degrees. In some embodiments, the preset threshold may be determined based on a system default setting or set manually by a user.
[0092] In some embodiments, the limited angular range may be determined based on a relative relationship between the radiation emitting assembly and the detector assembly. For example, when the direction of the radiation beam is perpendicular to the detector plane of the detector assembly 114 and faces the target breast, the angle between the radiation emitting assembly and the detector assembly may be determined as 0 degrees. If the limited angular range is from -25 degrees to 25 degrees, a sector region with a total scanning angle from 0 degrees to 50 degrees may be formed. If the limited angular range is from -25 degrees to 25 degrees, a sector region with a total scanning angle from 0 degrees to 50 degrees may be formed. If the limited angular range is from -25 degrees to -7.5 degrees and from 7.5 degrees to 25 degrees, a sector region with a total scanning angle from 15 degrees to 50 degrees may be formed.
[0093] In some embodiments, the plurality of projection angles may be uniformly disposed within the limited angular range. For example, within the limited angular range (e.g., from -25 degrees to 25 degrees) , a projection angle may be disposed every 1 degree, 1.5 degrees, 2 degrees, 2.5 degrees, 3 degrees, etc. Merely by way of example, the plurality of projection image sets may include a first projection image set corresponding to the first energy level (e.g., a low-energy level) and a second projection image set corresponding to the second energy level (e.g., a high-energy level) . The first projection image set may be collected at a plurality of projection angles (e.g., -25 degrees, -24 degrees, …, 24 degrees, 25 degrees) within the limited angular range at the first energy level, and the second projection image set may be collected at the plurality of projection angles (e.g., -25 degrees, -24 degrees, …, 24 degrees, 25 degrees) within the limited angular range at the second energy level.
[0094] In some embodiments, the plurality of projection image sets are collected by performing a multi-energy DBT scan using the multi-energy DBT device on the target breast. For example, if the multi-energy DBT device includes a radiation emitting assembly that can emit radiation beams at the energy levels, the radiation emitting assembly may rotate within the limited angular range and irradiate the target breast at different energy levels at each projection angle. As another example, if the multi-energy DBT device includes an energy-resolving detector, the radiation emitting assembly may rotate within the limited angular range and irradiate the target breast at a specific energy level at each projection angle, and the energy-resolving detector may resolve the radiation beams into different angle levels.
[0095] In some embodiments, before the multi-energy DBT scan, a contrast agent may be injected. Exemplary contrast agents may include an iodine contrast agent, a galactography contrast agent, or the like, or any combination thereof. For example, whether the injection of the contrast agent is needed may be determined based on a lesion to be examined. If the lesion to be examined is calcification, no contrast agent is needed, as the calcification is caused by metal deposition and does not require highlighting blood supply information for presentation. If the lesion to be examined is a soft tissue lesion, such as, a breast mass, the contrast agent needs to be injected to highlight blood supply information.
[0096] In some embodiments, the processing device 140 may obtain the plurality of projection image sets from the multi-energy DBT device or a storage device (e.g., the storage device 150, a database, or an external storage) that stores the plurality of projection image sets.
[0097] In 304, the processing device 140 (e.g., the generation module 220) may generate, based on the plurality of projection image sets, at least one 3D subtraction image of the target breast.
[0098] A 3D subtraction image refers to a 3D image obtained by subtracting images, in which background structures (e.g., a gland, fat tissues, etc. ) are partially or totally removed, and only regions of interest (ROI) are left. An ROI refers to a region that needs to be analyzed or observed for diagnosing the target breast and / or puncture guidance. For example, the ROI may include a boundary region of the target breast, a lesion region (e.g., tumor, calcification, the breast mass, etc. ) , etc.
[0099] In some embodiments, when the plurality of different energy levels include two energy levels, the at least one 3D subtraction image may include one 3D subtraction image of the target breast. When the plurality of different energy levels include at least three energy levels, the at least one 3D subtraction image may include one or more 3D subtraction images of the target breast, such as a 3D subtraction image corresponding to an energy level pair of any two energy levels among the at least three energy levels.
[0100] In some embodiments, the processing device 140 may generate at least one 2D subtraction image for each of the plurality of projection angles by performing a subtraction operation on the plurality of projection image sets, and then generate the at least one 3D subtraction image of the target breast based on the 2D subtraction images. Exemplary subtraction operations may include a multi-energy subtraction operation with energy switching, a multi-energy subtraction operation based on spectral detectors, a subtraction reconstruction based on time, a subtraction after reconstruction based on time, or the like, or any combination thereof. Merely by way of example, the plurality of energy levels may comprise a first energy level and a second energy level, and the projection image sets may comprise a first projection image set corresponding to the first energy level and a second projection image set corresponding to the second energy level. For each of the plurality of projection angles, the processing device 140 may generate a 2D subtraction image by performing dual-energy subtraction on a first projection image corresponding to the projection angle in the first projection image set and a second projection image corresponding to the projection angle in the second projection image set. The processing device 140 may generate the 3D subtraction image by reconstructing the plurality of 2D subtraction images corresponding to the plurality of projection angles. More descriptions regarding the generation of the 3D subtraction image may be found elsewhere in the present disclosure. See, e.g., FIG. 8 and relevant descriptions thereof.
[0101] In some embodiments, the processing device 140 may generate a preliminary 3D subtraction image for each of the plurality of energy levels, and generate the at least one 3D subtraction image of the target breast by performing the subtraction operation on the preliminary 3D subtraction images. Merely by way of example, the plurality of energy levels may comprise a first energy level and a second energy level, and the projection image sets may comprise a first projection image set corresponding to the first energy level and a second projection image set corresponding to the second energy level. The processing device 140 may generate a first 3D reconstruction image corresponding to the first energy level by reconstructing first projection images in the first projection image set, and generate a second 3D reconstruction image corresponding to the second energy level by reconstructing second projection images in the second projection image set. Further, the processing device 140 may generate the 3D subtraction image by performing dual-energy subtraction on the first 3D reconstruction image and the second 3D reconstruction image. More descriptions regarding the generation of the 3D subtraction image may be found elsewhere in the present disclosure. See, e.g., FIG. 9 and relevant descriptions thereof.
[0102] In some embodiments, artifact correction may be performed during the generation process of the at least one 3D subtraction image. For example, the artifact correction may be performed during the generation process via a constraint term configured to constrain a signal distribution of the ROI along a depth direction. More descriptions regarding the artifact correction may be found elsewhere in the present disclosure. See, e.g., FIGs. 8 and 9 and relevant descriptions thereof.
[0103] In some embodiments, the 3D subtraction image may be a 3D breast subtraction image. Merely by way of example, referring to FIG. 5, FIG. 5 is a schematic diagram illustrating an exemplary 3D subtraction image according to some embodiments of the present disclosure. As shown in FIG. 5, the processing device 140 may guide a puncture needle 520 to puncture an ROI 510 based on a 3D subtraction image 500 to obtain the living tissue for further diagnosis (e.g., tumor diagnosis) .
[0104] Through the subtraction operation, specific tissue structures in image data (e.g., the plurality of projection image sets, the preliminary 3D subtraction images, etc. ) can be extracted and enhanced, thereby highlighting these structures while weakening or removing other unnecessary information. For instance, through the subtraction operation, lesions (e.g., tumors in the 3D subtraction image) can be enhanced, while the interference from other tissue structures (e.g., fat tissues and blood vessels) can be weakened or removed, thereby improving the diagnostic effect (e.g., better guidance for puncture) .
[0105] According to some embodiments of the present disclosure, by introducing the multi-energy DBT technology, the radiation beams of different energy levels can be used to scan the target breast at a plurality of projection angles within a limited angular range, and the projection image sets of different layers of the target breast through multiple slices can be collected, thereby enhancing the contrast of breast tissues. As a result, the imaging results of a single scan can simultaneously provide 3D spatial information and functional information of the lesions, thereby improving the detection rate and accuracy of the lesion, and enhancing the accuracy of subsequent puncture guidance.
[0106] In 306, the processing device 140 (e.g., the generation module 220) may generate a rendered 3D subtraction image by rendering the 3D subtraction image.
[0107] A rendering operation refers to an operation of converting the 3D subtraction image into an ideal visualization image by processing the 3D subtraction image. For example, the rendered 3D subtraction image may include a 3D target breast model that is rotatable to display the target breast from different perspectives (e.g., the 3D target breast model can be rotated by a user to display the target breast from different perspectives) .
[0108] Through the rendering operation, the ROI in the 3D subtraction image can be highlighted, thereby enhancing the clarity, realism, and readability of the 3D subtraction image, and assisting the use in generating a diagnosis result of the target breast.
[0109] In some embodiments, the processing device 140 may preprocess the 3D subtraction image to generate a preprocessed 3D subtraction image, and generate the rendered 3D subtraction image by performing pseudo-color rendering on an ROI in the preprocessed 3D subtraction image.
[0110] In some embodiments, the preprocessing of the 3D subtraction image may comprise an image denoising operation, a boundary feature enhancement operation, an out-of-plane artifact reduction operation, an in-plane artifact reduction operation, or the like, or any combination thereof.
[0111] The image denoising operation is used to reduce noise (e.g., electronic noise, interference information, etc. ) in the 3D subtraction image. In some embodiments, the processing device 140 may remove the noise in the 3D subtraction image through a filtering technique, a deep learning-based image denoising technique (e.g., a convolutional neural network, a generative adversarial network) , etc.
[0112] A boundary feature refers to an image feature related to a boundary region (or edge) of the target breast in the 3D subtraction image. In some embodiments, the boundary feature may correspond to a boundary region between a human body region and a non-human body exposure region, which corresponds to skin edges of the human breast. The boundary feature enhancement operation is used to enhance the clarity of the boundary region.
[0113] In some embodiments, the processing device 140 may perform the boundary feature enhancement operation in various manners to enhance the boundary feature of the 3D subtraction image. For example, the processing device 140 may perform the boundary feature enhancement operation by adjusting the contrast of the 3D subtraction image through a contrast enhancement manner, thereby enhancing the contrast between the boundary region and the surrounding tissue. As another example, the processing device 140 may perform the boundary feature enhancement operation through an edge detection algorithm (e.g., a Canny edge detection algorithm, a Laplacian operator, etc. ) . As yet another example, the processing device 140 may perform the boundary feature enhancement operation by using an adaptive enhancement manner based on characteristics of the boundary region in the 3D subtraction image. For instance, the processing device 140 may adjust parameters (e.g., contrast, brightness, etc. ) of the 3D subtraction image based on information, such as, the gray value and a gradient value of the boundary region, making the breast edge more prominent. Herein, the boundary features may also be referred to as breast boundary features.
[0114] In some embodiments, the processing device 140 may also perform the boundary feature enhancement operation using a machine learning technique. For example, the processing device 140 may perform the boundary feature enhancement operation through a feature enhancement model. The feature enhancement model is a machine learning model. For example, the feature enhancement model may include a neural network model, a deep neural network model, a convolutional neural network model, or the like, or any combination thereof.
[0115] In some embodiments, the processing device 140 may input the 3D subtraction image into the feature enhancement model, and the feature enhancement model may output a 3D subtraction image with enhanced boundary features. The feature enhancement model may be obtained by performing model training based on first samples and corresponding first labels. The first samples may include a large number of historical 3D subtraction images, and the first labels may include a large number of historical 3D subtraction images with enhanced boundary features. The first samples and the first labels may be obtained through manual collection and annotation based on historical scans.
[0116] An out-of-plane artifact refers to information related to a focused layer that appears in a reconstruction image layer outside the focused layer due to incomplete image angles during the image reconstruction. FIG. 6A is a schematic diagram illustrating an exemplary non-focused layer of a subject during image reconstruction according to some embodiments of the present disclosure. FIG. 6B is a schematic diagram illustrating an exemplary focused layer of a subject during image reconstruction according to some embodiments of the present disclosure. The subject includes teel column beads. The focused layer refers to a specific imaging plane of the subject, and the non-focused layer refers to an image plane other than the specific imaging plane. As shown in FIGs. 6A and 6B, a subject 620 should appear in the focused layer as shown in FIG. 6B, but an out-of-plane artifact 610 corresponding to the subject appears in the non-focused layer as shown in FIG. 6A. It should be noted that the above descriptions regarding the subject and the out-of-plane artifact in FIGs. 6A and 6B are merely provided for the purposes of illustration, and not intended to limit the scope of the present disclosure.
[0117] Through the out-of-plane artifact reduction operation, the artifacts caused by the limited angular range can be reduced or eliminated.
[0118] An in-plane artifact refers to unexpected and unnecessary image information generated within the focused layer. For example, the in-plane artifact may include a motion artifact, a metal artifact, a stripe artifact, a noise artifact, a truncation artifact, or the like, or any combination thereof.
[0119] In some embodiments, the processing device 140 may preprocess the 3D subtraction image using a preprocessing model. In some embodiments, the preprocessing model is a machine learning model. For example, the preprocessing model may include a neural network model, a deep neural network model, a convolutional neural network model, or the like, or any combination thereof.
[0120] In some embodiments, the preprocessing model may include multiple layers. For example, the preprocessing model may include an image denoising layer, a boundary feature enhancement layer, an out-of-plane artifact reduction layer, an in-plane artifact reduction layer, or the like, or any combination thereof. In some embodiments, each layer of the preprocessing model may be trained separately or jointly. In some embodiments, the preprocessing model may be obtained by model training based on second samples and corresponding second labels. The second samples may include a large number of historical 3D subtraction images, and the second labels may include a large number of preprocessed historical 3D subtraction images. The second samples and the second labels may be obtained through manual collection and annotation based on the historical scans.
[0121] The pseudo-color rendering refers to applying colors (e.g., blue, green, red, etc. ) to the ROI to highlight the ROI.
[0122] In some embodiments, the processing device 140 may generate the rendered 3D subtraction image by performing the pseudo-color rendering on the ROI in the preprocessed 3D subtraction image. For example, the processing device 140 may perform, based on gray-scale information, the pseudo-color rendering on the ROI in the preprocessed 3D subtraction image.
[0123] By preprocessing the 3D subtraction image, the rendered 3D subtraction image can be generated by performing the pseudo-color rendering on the ROI in the preprocessed 3D subtraction image, thereby improving the display effects of the rendered 3D subtraction image when rendering and visualizing the 3D subtraction image. In this way, better guidance for puncture can be provided to the user (e.g., a doctor or a medical technician, etc. ) , thereby improving the accuracy of diagnosis.
[0124] In 308, the processing device 140 (e.g., the generation module 220) may post-process the rendered 3D subtraction image.
[0125] In some embodiments, the processing device 140 may visually display the ROI in the rendered 3D subtraction image. The visualization display refers to presenting the ROI to the user through a 3D model (e.g., the 3D target breast model as described in connection with 306) . For example, the processing device 140 may identify the ROI from the rendered 3D subtraction image, and add an annotation associated with the identified ROI on the rendered 3D subtraction image. Further, the processing device 140 may display, via a user interface (e.g., a user interface of the terminal device (s) 130) , the rendered 3D subtraction image with the annotation.
[0126] In some embodiments, the ROI may comprise the lesion region and / or the boundary region of the target breast, and the annotation associated with the identified ROI may be a bounding box containing the ROI or indicate a puncture route of the lesion region. For example, referring to FIG. 7, FIG. 7 is a schematic diagram illustrating an exemplary rendered 3D subtraction image seen from a certain perspective according to some embodiments of the present disclosure. As shown in FIG. 7, ROIs (e.g., a boundary region 710, a lesion region 720) may be identified from the rendered 3D subtraction image, and annotations may be added to the ROIs. For instance, a bounding box may be added to the lesion region 720.
[0127] In some embodiments, the processing device 140 may enhance the ROI in the rendered 3D subtraction image. For example, the processing device 140 may enhance the boundary region and the lesion region based on spatial positions of the boundary region and the lesion region in target breast. By enhancing the ROI, a user can easily identify the ROI while non-interesting feature information (e.g., the in-plane and out-of-plane artifacts, non-key breast fat, etc. ) can be eliminated or weakened.
[0128] In some embodiments, the visualization display of the ROI in the rendered 3D subtraction image may include displaying the ROI at any spatial angle by rotating the 3D target breast model in a 3D coordinate system, and displaying sectional information of the ROI in any direction such as a coronal plane, a sagittal plane, a transverse plane, etc. For example, the ROI in the 3D coordinate system as shown in FIG. 7 may be rotated to display features of the ROI at different perspectives.
[0129] In some embodiments, the processing device 140 may also guide the puncture needle to puncture the target breast based on the ROI in the rendered 3D subtraction image. For example, the puncture needle may be guided to puncture and collect tissues including the ROI (e.g., the lesion region) based on the spatial position of the ROI in the rendered 3D subtraction image. Meanwhile, a puncture depth and a puncture sampling process on the rendered 3D subtraction image may be intuitively presented through the rendered 3D subtraction image, which facilitates real-time adjustment of the puncture needle and improves the puncture accuracy, thereby accurately and conveniently completing the tissue sampling.
[0130] In some embodiments, by visualizing the ROI in the rendered 3D subtraction image, critical structures (e.g., a lesion structure) of the target breast may be displayed, thereby helping the user accurately identify and locate the position of the ROI.
[0131] In some embodiments, the processing device 140 may generate a diagnosis result of the target breast by analyzing the rendered 3D subtraction image. For example, the processing device 140 may assist in the determination of a malignancy degree of the target breast (e.g., the lesion region) based on the rendered 3D subtraction image in combination with a breast grading auxiliary diagnosis technology. The breast grading auxiliary diagnosis technology is used to detect the lesion region (e.g., the tumor) in the rendered 3D subtraction image (e.g., through a trained machine learning model) and grade the lesion region (e.g., negative, benign, suspicious malignant, malignant, etc. ) , thereby assisting the user in the determination of the malignancy degree of the lesion region.
[0132] According to some embodiments of the present disclosure, in the rendered 3D subtraction image, the ROI (e.g., the lesion region) can be intuitively displayed, so that the shape, size, benignity and malignancy, and even the blood flow and metabolic conditions of the ROI can be intuitively and accurately presented from different perspectives, thereby improving the efficiency and accuracy of the auxiliary diagnosis of the lesion region.
[0133] Furthermore, by combining the advantages of the multi-energy imaging and the DBT technology, imaging results obtained from a single scan can concurrently provide 3D spatial information and functional information of the ROI, thereby improving lesion detection rates and diagnostic accuracy, and further enhancing the precision of subsequent puncture guidance.
[0134] Moreover, the rendered 3D subtraction image undergoes the pseudo-color rendering and three-dimensional modeling, allowing arbitrary-angle rotation, multi-planar sectioning, and feature-enhancement display. Such processing facilitates clinical observation and patient comprehension, thereby improving the efficiency of doctor-patient communication.
[0135] In some embodiments, one or more 3D reconstruction images corresponding to one or more energy levels may be generated. For example, as described in connection with FIG. 9, a first 3D reconstruction image corresponding to the first energy level and a second 3D reconstruction image corresponding to the second energy level may be generated. The 3D subtraction image may be fused with a 3D reconstruction image, and the 3D fused image may then be post-processed and / or rendered.
[0136] FIG. 4 is a schematic diagram illustrating an exemplary process 400 for breast imaging according to some embodiments of the present disclosure.
[0137] As illustrated in FIG. 4, the processing device 140 may obtain a plurality of projection image sets 410 (e.g., a projection image set 402, a projection image set 404, etc. ) of a target breast collected using a multi-energy DBT device. Each of the plurality of projection image sets 410 may correspond to one of a plurality of different energy levels. For example, the plurality of energy levels may include a first energy level and a second energy level, the projection image set 402 may correspond to the first energy level, and the projection image set 404 may correspond to the second energy level.
[0138] The processing device 140 may generate at least one 3D subtraction image 420 of the target breast based on the plurality of projection image sets 410, and generate a rendered 3D subtraction image 430 by rendering the 3D subtraction image 420.
[0139] In some embodiments, the processing device 140 may identify an ROI 440 from the rendered 3D subtraction image 430, and add an annotation associated with the identified ROI 440 on the rendered 3D subtraction image 430 to generate a rendered 3D subtraction image 450 with the annotation. In 460, the processing device 140 may display, via a user interface, the rendered 3D subtraction image 450 with the annotation.
[0140] In some embodiments, the processing device 140 may further generate a diagnosis result 470 of the target breast by analyzing the rendered 3D subtraction image 430 and / or the rendered 3D subtraction image 450 with the annotation.
[0141] FIG. 8 is a schematic diagram illustrating an exemplary process 800 for generating a 3D subtraction image according to some embodiments of the present disclosure. FIG. 9 is a schematic diagram illustrating another exemplary process 900 for generating a 3D subtraction image according to some embodiments of the present disclosure.
[0142] As described in connection with FIG. 3, a plurality of projection image sets of a target breast may be collected using a multi-energy DBT device, each of the plurality of projection image sets may correspond to one of a plurality of different energy levels, and each projection image set may comprise a plurality of projection images that are collected at a plurality of projection angles within limited angular range at the corresponding energy level. In some embodiments, the sets of projection images corresponding to a pair of energy levels may be processed to generate a corresponding 3D subtraction image. The pair of energy levels may include any two different energy levels of the plurality of energy levels. For illustration purposes, it is assumed that the pair of energy levels includes a first energy level and a second energy level, the first energy level may be less than the second energy level, and the projection image sets may comprise a first projection image set corresponding to the first energy level and a second projection image set corresponding to the second energy level. The count of projection angles is denoted as N. Therefore, the projection angles include 802-1 to 802-N, the first projection image set includes first projection images 811-1 to 811-N, and the second projection image set includes second projection images 812-1 to 812-N, wherein the first projection image 811-i and the second projection image 812-i correspond to the projection angle 802-i.
[0143] In some embodiments, the 3D subtraction image is generated by performing the process 800. As shown in FIG. 8, for each projection angle 802, a 2D subtraction image 822 may be generated by performing dual-energy subtraction on a first projection image 811 corresponding to the projection angle 802 in the first projection image set and a second projection image 812 corresponding to the projection angle 802 in the second projection image set. Merely by way of example, as for the projection angle 802-1, the corresponding 2D subtraction image 822-1 is generated by performing dual-energy subtraction on the first projection image 811-1 and the second projection image 812-1. In this way, 2D subtraction image 822-1 to 822-N corresponding to the projection angles are obtained.
[0144] The dual-energy subtraction may be performed according to Equation (1) : SIDE=log (PHE) -wt×log (PLE) , (1) where SIDE represents a 2D subtraction image, PLE represents a first projection image corresponding to the first energy level (i.e., a low-energy level) , PHE represents a second projection image corresponding to the second energy level (i.e., a high-energy level) , and wt is a weighting factor (also referred to as a tissue cancellation factor or an energy weighting factor) .
[0145] wt may be configured to selectively eliminate or subtract a specific tissue (e.g., a glandular tissue) in the 2D subtraction image SIDE through precise mathematical calculations, thereby making another tissue (e.g., calcification / microcalcification) or a lesion region (e.g., tumor) more clearly visible. For example, by determining wt, a pixel value of a pixel of the specific tissue in the first projection image may be the same as or similar to a pixel value of a corresponding pixel in the second projection image. In some embodiments, wt may be determined based on a linear attenuation coefficient μ of the specific tissue (e.g., the glandular tissue) at a specific energy level. In some embodiments, wt may be preset in advance.
[0146] Furthermore, a 3D subtraction image 830 may be generated by reconstructing the 2D subtraction images 822-1 to 822-N. For example, the 3D subtraction image 830 may be generated by reconstructing the 2D subtraction images 822 through an image reconstruction algorithm. Exemplary image reconstruction algorithms may include a direct back projection algorithm, a filtered back projection algorithm, a convolutional back projection algorithm, a differential-Hilbert back projection algorithm, a gradient descent algorithm, an iterative reconstruction algorithm, or the like, or any combination thereof. As another example, the plurality of 2D subtraction images may be processed respectively using a contrast enhancement function to generate a plurality of enhanced 2D subtraction images, and the 3D subtraction image 830 may be generated by reconstructing the plurality of enhanced 2D subtraction images. An enhanced 2D subtraction image may be generated according to Equation (2) : SI′DE=f (SIDE) , (2) where f represents the contrast enhancement function, SI′DE represents the enhanced 2D subtraction image.
[0147] For illustration purposes, the generation (or reconstruction) process of the 3D subtraction image based on the 2D subtraction images is described below. In some embodiments, the plurality of 2D subtraction images corresponding to the plurality of projection angles may be reconstructed through an iterative reconstruction process, and artifact correction may be performed during the reconstruction process (e.g., the iterative reconstruction process) of the 3D subtraction image via a constraint term. The constraint term may be configured to constrain a signal distribution of an ROI along a depth direction. For instance, a current iteration in the iterative reconstruction process may include determining the signal distribution of the ROI along the depth direction in a current image to be updated in the current iteration, determining the constraint term based on the signal distribution, generating an updated image by updating the current image based on the plurality of 2D subtraction images and the signal distribution, and designating the updated image as a current image of a next iteration or designating the updated image as the 3D subtraction image.
[0148] The current image refers to an image to be updated in the current iteration. If the current iteration is a first iteration, the current image may be an original image. For example, the original image may be an image whose pixel values are 1. Alternatively, the original image may be an image obtained by reconstructing the plurality of 2D subtraction images through a filtered back projection algorithm.
[0149] The depth direction refers to a direction perpendicular to a detector plane of the detector assembly 114 and faces the target breast, and the depth direction is denoted as a Z direction. A horizontal direction and a longitudinal direction of the detector plane are respectively denoted as X and Y directions.
[0150] The signal distribution of the ROI along the depth direction may reflect variations in gray values of the ROI in the depth direction. The signal distribution of the ROI along the depth direction may be represented by a signal distribution curve. In some embodiments, the processing device 140 may identify the ROI from the current image, and then determine the signal distribution of the ROI along the depth direction.
[0151] In an ideal situation, the signal distribution curve of an ROI (e.g., a breast mass) with a clear boundary in the depth direction should have a sharp peak. However, due to artifacts, the signal distribution curve may have a short, wide, and trailing slope. Therefore, when the signal distribution curve has a short, wide, and trailing slope, the current image needs to be updated to reduce artifacts.
[0152] In some embodiments, the processing device 140 may determine a reconstruction target based on the signal distribution, and update the current image based on the plurality of 2D subtraction images and the reconstruction target. The reconstruction target may include a data fidelity term and the constraint term (also referred to as a regularization term) . The data fidelity term is associated with differences between forward projection of the current image and the plurality of 2D subtraction images (e.g., the 2D subtraction images 822-1 to 822-N) , and configured to make the reconstruction image as consistent as possible with the original projection images.
[0153] In some embodiments, the constraint term may be a correction function used to adjust the signal distribution of the ROI along the depth direction in the current image, to make it closer to the ideal signal distribution. For example, the constraint term may be constructed based on a total variation, L1 norm, L2 norm, etc. For instance, the constraint term may be a directional total variation as along one or more directions (e.g., the depth direction) . In some embodiments, the constraint term may include constraints on the signal distribution in different directions (e.g., the depth direction (the Z direction) , the horizontal direction (the X direction) , the longitudinal direction (the Y direction) ) , and the constraints in different directions may have different weights.
[0154] Merely by way of example, the reconstruction target may be represented by Equation (3) : where μ represents the 3D subtraction image to be reconstructed, represents the reconstruction target, D (μ) represents the data fidelity term, R (μ) represents the constraint term, and β represents a weight of the constraint term.
[0155] β is a hyperparameter greater than 0, used to control the weight of the constraint term R (μ) . The larger β is, the more the reconstruction result conforms to the prior assumption, but details may be overly smoothed. The smaller β is, the more faithful it may be to the measurement data (i.e., the original projection images) , but noise and artifacts may also be more obvious.
[0156] In some embodiments, the weight β of the constraint term may be dynamically adjusted according to the shape of the signal distribution curve (e.g., the severity of the tailing) . If the tailing is severe, the weight β or the weight of the constraint term in the Z direction in the current iteration may be increased to apply a relatively strong correction. As the iterations proceed, the signal distribution curve may become stable, and the weight β may be moderately reduced to prevent excessive smoothing.
[0157] In some embodiments, during the update of the current image, a data update step may be performed first to reduce the data fidelity term, and then the regularization constraint may be applied to the updated image according to the constraint term.
[0158] In some embodiments, the processing device 140 may determine whether a termination condition is met in the current iteration. The termination condition may include that a specific number of iterations is executed, the iteration converges, etc. For example, if a difference between the reconstruction results of the current iteration and the previous iteration is less than a first threshold, and a difference in the signal distribution of the ROI in the depth direction in the current iteration and the previous iteration is less than a second threshold, it is determined that the iteration converges. The first threshold and / or the second threshold may be determined based on a system default setting or set manually by a user. Through the iterative reconstruction process, the signal distribution of the ROI in the depth direction becomes increasingly stable and clear, and the updated image is consistent with the projection images.
[0159] In some embodiments, the processing device 140 may reconstruct a portion of the plurality of 2D subtraction images to generate the 3D subtraction image. For example, for each of the plurality of 2D subtraction images, the processing device 140 may segment the ROI from the 2D subtraction image to generate a local image of the ROI, and reconstruct the plurality of local images of the ROI to generate the 3D subtraction image. For instance, the processing device 140 may segment the ROI from the 2D subtraction image to generate the local image of the ROI through an image segmentation algorithm. Exemplary image segmentation algorithms may include a threshold-based segmentation algorithm, an edge-based segmentation algorithm, a cluster analysis-based segmentation algorithm, a wavelet transform-based segmentation algorithm, a neural network-based segmentation algorithm, or the like, or any combination thereof. In some embodiments, the plurality of local images of the ROI may be reconstructed in a similar manner as how the 2D subtraction images 822-1 to 822-N are reconstructed.
[0160] According to some embodiments of the present disclosure, by introducing the constraint term in the depth direction into the iterative reconstruction process, the depth artifacts caused by the limited angular range are specifically suppressed, thereby significantly improving the image resolution and the clarity of lesion boundaries. Furthermore, by performing the dual-energy subtraction before the reconstruction, only one reconstruction operation is needed, thereby reducing the computational load and improving the efficiency of the image reconstruction.
[0161] In some embodiments, the 3D subtraction image is generated by performing the process 900. As shown in FIG. 9, a first 3D reconstruction image 922 corresponding to the first energy level may be generated by reconstructing the first projection images 811-1 to 811-N in the first projection image set. The first 3D reconstruction image 922 may be reconstructed in a similar manner as how the 3D subtraction image 830 is generated as described in FIG. 8.
[0162] Similarly, a second 3D reconstruction image 924 corresponding to the second energy level may be generated by reconstructing the second projection images 812-1 to 812-N in the second projection image set. The second 3D reconstruction image 924 may be reconstructed in a similar manner as how the 3D subtraction image 830 is generated as described in FIG. 8.
[0163] Furthermore, a 3D subtraction image 930 may be generated by performing dual-energy subtraction on the first 3D reconstruction image 922 and the second 3D reconstruction image 924. The dual-energy subtraction may be performed according to Equation (4) : VDE=VHE-wt×VLE, (4) where VDE represents the 3D subtraction image 930, VLE represents the first 3D reconstruction image 922, and VHE represents the second 3D reconstruction image 924.
[0164] In some embodiments, the 3D subtraction image 930 may be further processed using a contrast enhancement function to generate an enhanced 3D subtraction image 940. For example, the enhanced 3D subtraction image 940 may be generated according to Equation (5) : V′DE=f (VDE) , (5) where V′DE represents the enhanced 3D subtraction image 940.
[0165] FIG. 10 is a flowchart illustrating an exemplary process 1000 for generating a target 3D subtraction image according to some embodiments of the present disclosure. The process 1000 may be performed when there are at least three energy levels.
[0166] In 1002, the processing device 140 (e.g., the generation module 220) may determine a characteristic energy level of an ROI of a target breast.
[0167] The characteristic energy level refers to an energy level at which the X-ray attenuation properties of the ROI exhibit a dramatic shift. There is a significant difference in the ROI’s attenuation characteristic of the radiation beams before and after this energy level. For example, the ability of the iodine substance to absorb radiation beams before 33.2 keV is relatively weak, and the ability of the iodine substance to absorb radiation beams after 33.2 keV suddenly increases. Therefore, the characteristic energy level of the iodine substance is 33.2 keV. For mass lesions, such as tumors, which may not have an inherent characteristic energy level, a contrast agent like iodine is often administered. In such cases, the characteristic energy level of the mass lesion is defined by the characteristic energy level of the contrast agent (e.g., 33.2 keV for iodine) . As another example, since calcium metal inherent in calcification has a distinct abrupt change in the attenuation characteristics of radiation beams at 40 keV, the characteristic energy level of the calcification may be 40 keV.
[0168] In 1004, the processing device 140 (e.g., the generation module 220) may determine, based on the characteristic energy level, two target energy levels from at least three energy levels.
[0169] The two target energy levels are selected to straddle the characteristic energy level, that is, one of the two target energy levels may be larger than the characteristic energy level, and the other one of the two target energy levels may be less than the characteristic energy level. This selection ensures that the ROI exhibits a substantial difference in attenuation between the two projection image sets corresponding to the two target energy levels, which is amplified during the subtraction process. Consequently, the resulting target 3D subtraction image provides enhanced visual contrast of the ROI relative to the surrounding tissue.
[0170] In 1006, the processing device 140 (e.g., the generation module 220) may generate a target 3D subtraction image for the ROI based on projection image sets corresponding to the two target energy levels.
[0171] The target 3D subtraction image refers to a 3D subtraction image most suitable for the ROI. In some embodiments, the processing device 140 may determine the projection image sets corresponding to the two target energy levels, and generate the target 3D subtraction image for the ROI. The target 3D subtraction image may be generated in a similar manner as how the 3D subtraction image is generated as described in FIGs. 8 and 9.
[0172] Merely by way of example, when the detector assembly 114 is a photon-counting flat panel detector, a plurality of energy levels comprise at least three energy levels (e.g., 20 keV, 30 keV, and 40 keV) , and the characteristic energy level of the ROI is 33.2 keV. The processing device 140 may determine 30 keV and 40 keV as the two target energy levels from 20 keV, 30 keV, and 40 keV based on the characteristic energy level of 33.2 keV, and generate the target 3D subtraction image for the ROI based on projection image sets corresponding to 30 keV and 40 keV.
[0173] Through the analysis of the characteristic energy level, the target 3D subtraction image that is most suitable for the ROI can be generated, thereby improving the accuracy of the target 3D subtraction image and providing more accurate information to the user, thus improving lesion detection rates and diagnostic accuracy.
[0174] FIG. 11 is a flowchart illustrating an exemplary process 1100 for generating a 3D subtraction image according to some embodiments of the present disclosure.
[0175] As shown in FIG. 11, when the detector assembly 114 is a photon-counting flat panel detector, a plurality of energy levels comprise at least three energy levels (e.g., a first energy level 1102, a second energy level 1104, a third energy level 1106, etc. ) .
[0176] The processing device 140 may determine a plurality of energy level pairs among the at least three energy levels. An energy level pair refers to a pair of two energy levels selected from the at least three energy levels. For example, as shown in FIG. 11, the plurality of energy level pairs may include a first energy level pair of the first energy level 1102 and the second energy level 1104, a second energy level pair of the first energy level 1102 and the third energy level 1106, and a third energy level pair of the second energy level 1104 and the third energy level 1106.
[0177] For each energy level pair, the processing device 140 may generate a 3D subtraction image based on projection image sets corresponding to the energy level pairs. For example, a 3D subtraction image 1122 may be generated based on projection image sets 1112 and 1114 corresponding to the first energy level pair, a 3D subtraction image 1124 may be generated based on the projection image set 1112 and a projection image set 1116 corresponding to the second energy level pair, a 3D subtraction image 1126 may be generated based on the projection image sets 1114 and 1116 corresponding to the third energy level pair.
[0178] In some embodiments, the subtraction images 1122, 1124, and 1126 can be presented to the user, so that the user can understand the condition of the target breast from different perspectives.
[0179] In some embodiments, for each energy level pair, the processing device 140 may generate an identification result of an ROI by identifying the ROI from the corresponding 3D subtraction image, and select a target 3D subtraction image 1130 corresponding to the ROI from the plurality of 3D subtraction images corresponding to the energy level pairs based on the identification results corresponding to the plurality of energy level pairs. For example, a 3D subtraction image with the greatest image quality (e.g., the lowest degree of artifacts) for the ROI is selected as the target 3D subtraction image 1130.
[0180] In some embodiments, a plurality of target 3D subtraction images may be generated for a plurality of ROIs of the target breast, and a fused 3D subtraction image may be generated by fusing the plurality of target 3D subtraction images of the plurality of ROIs. The fused 3D subtraction image may have great image quality for the plurality of ROIs. For example, the processing device 140 may register the plurality of target 3D subtraction images, and generate the fused 3D subtraction image by fusing the registered target 3D subtraction images.
[0181] The presentation of multiple ROIs within the fused 3D subtraction image enables the user to interpret the condition of all ROIs in a single review, thereby enhancing diagnostic efficiency.
[0182] In some embodiments, after a plurality of projection image sets corresponding to the plurality of energy levels are collected, a 3D reconstruction image (e.g., the first 3D reconstruction image and the second 3D reconstruction image as described in connection with FIG. 9) corresponding to each energy level is generated based on the corresponding projection image set. The 3D reconstruction images of the energy levels are rendered to generate 3D models corresponding to the energy levels. These 3D models are subsequently fused to produce a composite 3D fused model. The composite 3D fused model, integrating multi-energy information, is then utilized for precise needle guidance and targeting.
[0183] It should be noted that the descriptions of the processes 300, 400, and 800-1100 are provided for the purposes of illustration, and are not intended to limit the scope of the present disclosure. For persons having ordinary skills in the art, various variations and modifications may be conducted under the teaching of the present disclosure. For example, the processes 300, 400, and 800-1100 may be accomplished with one or more additional operations not described, and / or without one or more of the operations discussed. Additionally, the order in which the operations of the processes 300, 400, and 800-1100 are not intended to be limiting. However, those variations and modifications may not depart from the protection of the present disclosure.
[0184] Having thus described the basic concepts, it may be rather apparent to those skilled in the art after reading this detailed disclosure that the foregoing detailed disclosure is intended to be presented by way of example only and is not limiting. Although not explicitly stated here, those skilled in the art may make various modifications, improvements, and amendments to the present disclosure. These alterations, improvements, and modifications are intended to be suggested by this disclosure and are within the spirit and scope of the exemplary embodiments of this disclosure.
[0185] Moreover, certain terminology has been used to describe embodiments of the present disclosure. For example, the terms “one embodiment, ” “an embodiment, ” and / or “some embodiments” mean that a particular feature, structure, or feature described in connection with the embodiment is included in at least one embodiment of the present disclosure. Therefore, it is emphasized and should be appreciated that two or more references to “an embodiment” or “one embodiment” or “an alternative embodiment” in various portions of the present disclosure are not necessarily all referring to the same embodiment. In addition, some features, structures, or characteristics of one or more embodiments in the present disclosure may be properly combined.
[0186] Furthermore, the recited order of processing elements or sequences, or the use of numbers, letters, or other designations therefore, is not intended to limit the claimed processes and methods to any order except as may be specified in the claims. Although the above disclosure discusses some embodiments of the invention currently considered useful by various examples, it should be understood that such details are for illustrative purposes only, and the additional claims are not limited to the disclosed embodiments. Instead, the claims are intended to cover all combinations of corrections and equivalents consistent with the substance and scope of the embodiments of the invention. For example, although the implementation of various components described above may be embodied in a hardware device, it may also be implemented as a software only solution, e.g., an installation on an existing server or mobile device.
[0187] Similarly, it should be appreciated that in the foregoing description of embodiments of the present disclosure, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the various embodiments. However, this disclosure does not mean that object of the present disclosure requires more features than the features mentioned in the claims. Rather, claimed subject matter may lie in less than all features of a single foregoing disclosed embodiment.
[0188] In some embodiments, the numbers expressing quantities or properties used to describe and claim certain embodiments of the present disclosure are to be understood as being modified in some instances by the term “about, ” “approximate, ” or “substantially. ” For example, “about, ” “approximate, ” or “substantially” may indicate ±20%variation of the value it describes, unless otherwise stated. Accordingly, in some embodiments, the numerical parameters set forth in the written description and attached claims are approximations that may vary depending upon the desired properties sought to be obtained by a particular embodiment. In some embodiments, the numerical parameters should be construed in light of the number of reported significant digits and by applying ordinary rounding techniques. Notwithstanding that the numerical ranges and parameters setting forth the broad scope of some embodiments of the present disclosure are approximations, the numerical values set forth in the specific examples are reported as precisely as practicable.
[0189] Each of the patents, patent applications, publications of patent applications, and other material, such as articles, books, specifications, publications, documents, things, and / or the like, referenced herein is hereby incorporated herein by this reference in its entirety for all purposes. History application documents that are inconsistent or conflictive with the contents of the present disclosure are excluded, as well as documents (currently or subsequently appended to the present specification) limiting the broadest scope of the claims of the present disclosure. By way of example, should there be any inconsistency or conflict between the description, definition, and / or the use of a term associated with any of the incorporated material and that associated with the present document, the description, definition, and / or the use of the term in the present document shall prevail.
[0190] In closing, it is to be understood that the embodiments of the present disclosure disclosed herein are illustrative of the principles of the embodiments of the present disclosure. Other modifications that may be employed may be within the scope of the present disclosure. Thus, by way of example, but not of limitation, alternative configurations of the embodiments of the present disclosure may be utilized in accordance with the teachings herein. Accordingly, embodiments of the present disclosure are not limited to that precisely as shown and described.
Claims
A method for breast imaging, implemented on a computing device having at least one processor and at least one storage device, the method comprising:obtaining a plurality of projection image sets of a target breast collected using a multi-energy digital breast tomosynthesis (DBT) device, wherein each of the plurality of projection image sets corresponds to one of a plurality of energy levels, each projection image set comprises a plurality of projection images that are collected at a plurality of projection angles;generating, based on the plurality of projection image sets, at least one three-dimensional (3D) subtraction image of the target breast, wherein artifact correction is performed during the generation process of the at least one 3D subtraction image; andgenerating a rendered 3D subtraction image by rendering the 3D subtraction image.The method of claim 1, further comprising:identifying a region of interest (ROI) from the rendered 3D subtraction image;adding an annotation associated with the identified ROI on the rendered 3D subtraction image; anddisplaying, via a user interface, the rendered 3D subtraction image with the annotation.The method of claim 2, wherein the ROI comprises a lesion region, and the annotation associated with the identified ROI indicates a puncture route of the lesion region.The method of any one of claims 1-3, further comprising:generating a diagnosis result of the target breast by analyzing the rendered 3D subtraction image.The method of any one of claims 1-4, wherein the rendered 3D subtraction image includes a 3D target breast model that is rotatable to display the target breast from different perspectives.The method of any one of claims 1-5, wherein the generating a rendered 3D subtraction image by rendering the 3D subtraction image comprises:preprocessing the 3D subtraction image to generate a preprocessed 3D subtraction image; andgenerating the rendered 3D subtraction image by performing pseudo-color rendering on an ROI in the preprocessed 3D subtraction image.The method of any one of claims 1-6, wherein each of the plurality of energy levels is within a range of 20 keV to 50 keV.The method of any one of claims 1-7, wherein the plurality of projection angles are within a limited angular range, the span of the limited angular range is not greater than 50 degrees.The method of any one of claims 1-8, wherein the plurality of energy levels comprise a first energy level and a second energy level, the projection image sets comprise a first projection image set corresponding to the first energy level and a second projection image set corresponding to the second energy level, and the generating at least one 3D subtraction image of the target breast comprises:for each of the plurality of projection angles, generating a two-dimensional (2D) subtraction image by performing dual-energy subtraction on a first projection image corresponding to the projection angle in the first projection image set and a second projection image corresponding to the projection angle in the second projection image set;generating a 3D subtraction image by reconstructing the plurality of 2D subtraction images corresponding to the plurality of projection angles.The method of claim 9, wherein the artifact correction is performed during the generation process of the 3D subtraction image via a constraint term configured to constrain a signal distribution of an ROI along a depth direction.The method of claim 10, wherein the plurality of 2D subtraction images corresponding to the plurality of projection angles are reconstructed through an iterative reconstruction process, and a current iteration in the iterative reconstruction process comprises:determining the signal distribution of the ROI along the depth direction in a current image to be updated in the current iteration;determining the constraint term based on the signal distribution;generating an updated image by updating the current image based on the plurality of 2D subtraction images and the constraint term; anddesignating the updated image as a current image of a next iteration, or designating the updated image as the 3D subtraction image.The method of claim 11, wherein the updating the current image is performed based on a reconstruction target comprising a data fidelity term and the constraint term, the data fidelity term being associated with a difference between forward projection of the current image and the plurality of 2D subtraction images.The method of claim 10, wherein the constraint term is a directional total variation.The method of claim 9, wherein the generating a 3D subtraction image by reconstructing the plurality of 2D subtraction images corresponding to the plurality of projection angles comprises:for each of the plurality of 2D subtraction images, segmenting an ROI from the 2D subtraction image to generate a local image of the ROI; andreconstructing the plurality of local images of the ROI to generate the 3D subtraction image.The method of any one of claims 1-8, wherein the plurality of energy levels comprise a first energy level and a second energy level, the projection image sets comprise a first projection image set corresponding to the first energy level and a second projection image set corresponding to the second energy level, and the generating at least one 3D subtraction image of the target breast comprises:generating a first 3D reconstruction image corresponding to the first energy level by reconstructing first projection images in the first projection image set;generating a second 3D reconstruction image corresponding to the second energy level by reconstructing second projection images in the second projection image set; andgenerating the 3D subtraction image by performing dual-energy subtraction on the first 3D reconstruction image and the second 3D reconstruction image.The method of any one of claims 1-15, wherein the multi-energy DBT device comprises a radiation source that is capable of switching between the plurality of energy levels, or the multi-energy DBT device comprises a plurality of radiation sources respectively corresponding to the plurality of energy levels.The method of any one of claims 1-15, wherein the multi-energy DBT device comprises an energy-resolving detector that is capable of distinguishing radiation at the plurality of energy levels.The method of claim 17, wherein the energy-resolving detector is a photon-counting flat panel detector, the plurality of energy levels comprise at least three energy levels, and the generating, based on the plurality of projection image sets, at least one 3D subtraction image of the target breast comprises:determining a characteristic energy level of an ROI of the target breast;determining, based on the characteristic energy levels, two target energy levels from the at least three energy levels; andgenerating a target 3D subtraction image for the ROI based on the projection image sets corresponding to the two target energy levels.The method of claim 17, wherein the energy-resolving detector is a photon-counting flat panel detector, the plurality of energy levels comprise at least three energy levels, and the generating, based on the plurality of projection image sets, at least one 3D subtraction image of the target breast comprises:determining a plurality of energy level pairs among the at least three energy levels; andfor each energy level pair, generating a 3D subtraction image based on the projection image sets corresponding to the energy level pairs.A system for breast imaging, comprising:at least one storage device including a set of instructions; andat least one processor configured to communicate with the at least one storage device, wherein when executing the set of instructions, the at least one processor is configured to direct the system to perform operations including:obtaining a plurality of projection image sets of a target breast collected using a multi-energy digital breast tomosynthesis (DBT) device, wherein each of the plurality of projection image sets corresponds to one of a plurality of energy levels, each projection image set comprises a plurality of projection images that are collected at a plurality of projection angles;generating, based on the plurality of projection image sets, at least one three-dimensional (3D) subtraction image of the target breast, wherein artifact correction is performed during the generation process of the at least one 3D subtraction image; andgenerating a rendered 3D subtraction image by rendering the 3D subtraction image.A non-transitory computer readable medium, comprising executable instructions that, when executed by at least one processor, direct the at least one processor to perform a method, the method comprising:obtaining a plurality of projection image sets of a target breast collected using a multi-energy digital breast tomosynthesis (DBT) device, wherein each of the plurality of projection image sets corresponds to one of a plurality of energy levels, each projection image set comprises a plurality of projection images that are collected at a plurality of projection angles;generating, based on the plurality of projection image sets, at least one three-dimensional (3D) subtraction image of the target breast, wherein artifact correction is performed during the generation process of the at least one 3D subtraction image; andgenerating a rendered 3D subtraction image by rendering the 3D subtraction image.