Mammography imaging system with enhanced image resolution and method of use

Abstract

An imaging device or system (10) (e.g., a mammography imaging system (12)) and associated method that provides artifact-reduced images (420, 422, 424) of a subject (52). The system (10) and method generate a standard reconstructed volume (410) of the subject (52) created from one or more projection images (404) of the subject obtained over an angular range of less than + / - 60 degrees relative to the subject (52). A resolution-enhancing artificial intelligence (414) is applied to the projection images (404) to form an artifact-reduced / quasi-isotropic volume (416), and one or more quasi-isotropic images (420, 422, 424) are obtained therefrom along a plane normal to the plane of the detector (18) for the mammography imaging system (12). The resolution of the quasi-isotropic images (420, 422, 424) is similar to images obtained in a simulated computed tomography imaging procedure, and the quasi-isotropic volume (416) can be used to provide effective diagnostic images using strain elastography and digital decompression of the quasi-isotropic volume (416) to create a digital uncompressed subject volume (800) that can be modified.

Description

Technical Field

[0001] This disclosure relates generally to medical imaging systems, including mammography systems and devices, and more specifically to generating and utilizing enhanced resolution images produced by mammography systems. Background Technology

[0002] Embodiments of the present invention relate generally to X-ray medical imaging, and more specifically to devices, systems and methods for performing various imaging procedures such as mammography procedures, including but not limited to digital breast tomography (DBT) mammography, spectral mammography (SM), such as 2D / 3D dual-energy contrast-enhanced (CE) mammography or full-field digital mammography (FFDM).

[0003] For breast imaging, some exemplary imaging procedures include full-field digital mammography, which captures images directly onto a flat panel detector, and computed tomography involving the use of a cassette including an imaging plate. Alternatively, spectral mammography (SM), an X-ray imaging modality used to scan the breast for screening, diagnosis, and / or interventional procedures, can be employed. However, the effectiveness of these types of mammographic imaging is affected by numerous factors, one of which is the two-dimensional (2D) rendering of the images obtained using these types of mammographic imaging.

[0004] As an improvement over the aforementioned mammographic imaging process that only produces 2D images of the breast, the Digital Breast Tomography (DBT) system is a dedicated mammography system that can acquire several (e.g., dozens and / or between 9 and 40) angularly offset projected X-ray images within a limited angular range relative to the breast. The DBT system can use the resulting X-ray image data to reconstruct a three-dimensional (3D) image dataset that shows the entire volume of the breast differing in a direction orthogonal to the detector plane (“z-direction”).

[0005] 3D image datasets are used to form various volumetric representations of the imaged breast, including the entire 3D volume of the breast and various 3D segments of that volume, such as tomographic planes or slices of predefined thickness oriented to provide a desired view of one or more regions of interest (ROIs) detected within the 3D image dataset. However, these tomographic planes are obtained from a limited angle of acquisition projection and include undesirable information from regions above and below the generated tomographic planes without hard constraints, resulting in indistinct or unclear boundaries of the tomographic planes from the 3D image dataset. Due to the limited angular range of tomographic fusion acquisition, the volume is presented in a highly anisotropic representation, where the pixel pitch / spatial resolution in the axial plane is on the same order of magnitude as the pixel pitch / spatial resolution in the detector and an order of magnitude smaller than the pixel pitch / spatial resolution in the orthogonal direction.

[0006] Additionally, once a 3D image dataset of the breast has been generated, this dataset can be used, after being used in appropriate diagnostic procedures, to guide a biopsy device used with a DBT system into the breast to obtain a biopsy of a region of interest (ROI) identified within the 3D image dataset. In the DBT system, the biopsy device is directly mounted on the system to enable biopsies to be performed using the 3D image dataset or stereo camera images of the breast and the biopsy device. The biopsy device is then triangulated to the ROI within the breast to guide it to that ROI.

[0007] Regarding the use of these DBT mammography systems, the system setup for acquiring images requires attaching various devices to the system to provide proper breast positioning (i.e., compression) to obtain the desired image quality. In a mammography system, devices attached to the system to perform imaging and / or biopsy procedures include a compression paddle, a magnifying device, and / or a biopsy holder used to position the biopsy device on the mammography system in a location where the biopsy device can perform the desired biopsy procedure under the guidance of the mammography system. When the mammography imaging system operates in a screening configuration, the compression paddle and optionally the magnifying device (or magnetic support) are connected to the system. Conversely, when the imaging system is used in a diagnostic configuration, the compression paddle and biopsy positioner or holder, optionally along with the magnifying device and compatible with the operation of the biopsy device on the biopsy holder, are secured to the imaging system. In both configurations, the detector is fixed as part of the imaging system or can be rotated to an angle following the source within a fixed breast support, wherein a bucky is secured to the detector to provide a suitable X-ray transparent breast support surface and an image-enhancing anti-scattering grid located within the bucky.

[0008] In existing diagnostic mammography imaging devices, such as the DBT system, the radiation source is positioned directly above the detector, and the object being imaged (e.g., the breast) is positioned at a compression site adjacent to the detector. In this configuration, X-rays emitted by the radiation source pass through the breast along an axis defined between the radiation source and the detector to generate a 2D projected image of the breast. Furthermore, the radiation source can be moved or rotated to different angular positions relative to the compressed breast and the detector to generate additional 2D projected images used by the mammography imaging device / DBT system to reconstruct the 3D volume of the breast.

[0009] However, due to the geometry of the DBT system, i.e., the limited angular range, strong artifacts appear in the 3D volume reconstructed from 2D X-ray projections acquired along specific planes. Most notably, the resolution is severely reduced along the vertical detector-to-source axis (e.g., along the coronal, sagittal, and / or any other plane orthogonal to the detector or along any line between the detector and the focal point of the radiation source), and breast structures superimposed in this direction cannot be clearly separated. Figure 1 and Figure 1A The figure illustrates an example of the current state of technology for reprojected images obtained from reconstructed 3D volumes along coronal, sagittal, and transverse or axial planes. As shown, only images along the transverse plane (i.e., the plane perpendicular to the vertical detector-to-source axis) provide useful diagnostic information. Therefore, radiologists cannot currently utilize information from all line-of-sight directions from the 3D volume because only one of these directions is usable due to the geometric orientation of artifacts. In particular, diagnostically acceptable tomographic image planes cannot be generated except for those planes clearly parallel to the detector orientation, and even these image planes are "contaminated" by images of large structures present in adjacent transverse planes above and / or below the tomographic image plane.

[0010] Over the past few years, advancements in AI-based reconstruction algorithms have enabled the generation of reconstructed volumes with improved image quality along the vertical source-to-detector axis. However, research into imaging system design, image review methods, and applications leveraging these advancements has yet to commence.

[0011] Therefore, in view of the aforementioned drawbacks of existing mammography imaging systems regarding image quality along the vertical source-to-detector axis, it is desirable to develop a mammography system and associated method for generating improved reprojected images along the vertical source-to-detector plane to provide enhanced image information for use in diagnostic and treatment procedures. Summary of the Invention

[0012] According to one aspect of an exemplary embodiment of the present disclosure, a mammography imaging system includes: a radiation source operable to emit radiation; a detector alignable with the radiation source and having a surface on which a breast to be imaged is adapted to be positioned; a controller operably connected to the radiation source and the detector to control the operation of the radiation source and the detector, thereby generating image data of the breast in an imaging procedure executed by the imaging system, the controller including a central processing unit and an interconnected database including processor-executable instructions for processing image data from the detector to create one or more projected images of the breast; a display operably connected to the controller to present information to a user; and a user interface operably connected to the controller to enable user input to the controller, wherein the controller is configured to apply resolution-enhancing artificial intelligence to one or more projected images to reconstruct a quasi-isotropic volume of the breast, and to generate one or more quasi-isotropic images from the quasi-isotropic volume.

[0013] According to another aspect of an exemplary embodiment of the present disclosure, a method is used to provide one or more projected images of an object obtained in an angular range of less than 180 degrees relative to the object, to reconstruct a quasi-isotropic volume from the one or more projected images using artifact reduction and resolution enhancement artificial intelligence, and to generate one or more quasi-isotropic images from the quasi-isotropic volume.

[0014] According to another aspect of an exemplary embodiment of this disclosure, a method for providing an image of an object with reduced artifacts includes the steps of: providing a mammography imaging system comprising: a radiation source operable to emit radiation; a detector alignable to the radiation source and having a surface on which a breast to be imaged is adapted to be positioned; a controller operably connected to the radiation source and the detector to control the operation of the radiation source and the detector, thereby generating image data of the breast in an imaging procedure executed by the imaging system, the controller including a central processing unit and an interconnected database including processor-executable instructions for processing the image data from the detector to create one or more projected images; a display operably connected to the controller to present information to a user; and a user interface operably connected to the controller to enable user input to the controller; placing the breast on the surface of the detector; operating the radiation source within a limited angular range relative to the breast to obtain image data; processing the image data to form one or more projected images; reconstructing a quasi-isotropic volume from the one or more projected images using artifact reduction and resolution enhancement artificial intelligence; and generating one or more quasi-isotropic images from the quasi-isotropic volume.

[0015] These and other exemplary aspects, features and advantages of the invention will become apparent from the following detailed description taken in conjunction with the accompanying drawings. Attached Figure Description

[0016] The accompanying drawings illustrate the currently conceived best mode for practicing the present invention.

[0017] In the attached diagram: Figure 1 and Figure 1A It is a schematic diagram of various prior art images sampled along various axes of the image volume produced by a mammography imaging system of current technology, and axis references used for various images.

[0018] Figure 2 This is a perspective view of an imaging device in the form of a mammography system for imaging a patient's breast tissue, according to an embodiment of this disclosure.

[0019] Figure 3 yes Figure 2 The diagram shows the radiation source of a mammography system in the scanning position according to an embodiment of the present disclosure.

[0020] Figure 4 This is an operation based on the implementation scheme of this disclosure. Figure 2 A flowchart of the steps of an exemplary method for using a mammography system to produce 3D volumes with reduced artifacts.

[0021] Figure 5 This is a schematic diagram of a display according to an embodiment of the present disclosure, the display including a cross-sectional view obtained from a standard reconstructed volume, and a view presented therewith, via... Figure 4 The method generates artifacts that reduce volume to obtain coronal and sagittal views.

[0022] Figure 6 This is a schematic diagram of various synthetic images obtained along various planes in a simulated computed tomography (CT) imaging procedure of a breast that is compressed and positioned as in a standard DBT system, according to embodiments of the present disclosure, to produce Figure 5 The image in the image is as Figure 5 A reference for the accuracy of the image.

[0023] Figure 7 It is based on the implementation scheme of this disclosure. Figure 4 The method generates artifact reduction volumetric images sampled along the transverse, coronal, and sagittal planes.

[0024] Figure 8 A and Figure 8 B is the operation according to the implementation scheme of this disclosure. Figure 4The method is to reduce the volume of output thick slices from artifacts. Figure 8 A) and synthetic 2D images ( Figure 8 A schematic diagram of method B).

[0025] Figure 9 It is based on the implementation scheme of this disclosure. Figure 4 The volumetric image obtained by reducing the volume of artifacts generated by the method.

[0026] Figure 10 It is a prior art display for planning biopsy procedures, including a cross-sectional view obtained from a standard reconstructed volume indicating the biopsy target, which is presented as a schematic diagram of the breast to be biopsied and a representation of the biopsy needle positioned to perform the biopsy procedure.

[0027] Figure 11 The display, set according to an embodiment of this disclosure, is used for planning a biopsy procedure, including a cross-sectional view obtained from a standard reconstructed volume, indicating the biopsy target, which is presented as having [the following information is missing from the original text] Figure 4 The method produces a sagittal view of the breast being biopsied by reducing volume through artifacts, and a representation of the biopsy needle positioned to perform the biopsy procedure.

[0028] Figure 12 The embodiments of the present disclosure are shown by Figure 4 The method generates artifacts that reduce the volume of cross-sectional and coronal views, and represents the 3D ellipsoidal annotations applied to the cross-sectional and coronal views.

[0029] Figure 13 It is based on the implementation scheme of this disclosure. Figure 4 A schematic diagram of a cross-sectional view of the breast obtained by reducing the volume of artifacts produced by the method, and a representation of the uncertainty diagram and fraction of the cross-sectional view.

[0030] Figures 14A to 14D The implementation of the present disclosure is shown by means of... Figure 4 The method produces artifacts that reduce volume. Mechanical numerical simulations are used to obtain various schematic diagrams of simulated compression on uncompressed breasts and simulated decompression on compressed breasts.

[0031] Figures 15A to 15C An example is illustrated in the embodiment of this disclosure, from the reconstructed volume (a), by Figure 4 The image shows the result of tissue density estimation obtained from the artifact reduction volume generated by the method and the material composition image derived from the artifact reduction volume (c).

[0032] Figure 16 It is based on the implementation scheme of this disclosure. Figure 2A schematic diagram of a mammography imaging system used to perform X-ray strain elastography procedures.

[0033] Figure 17 This is an operation based on the implementation scheme of this disclosure. Figure 2 A flowchart of the steps of an exemplary method for using a mammography system to perform an X-ray strain elastography procedure.

[0034] Figures 18A to 18B This illustrates the angular coverage of X-ray acquisition for spherical objects and is derived from... Figure 4 The diagram illustrates the modifications produced by the method, which aim to reduce volume in generating artifacts with quasi-isotropic resolution. Detailed Implementation

[0035] One or more specific implementations will be described below. To provide a concise description of these implementations, not all characteristics of the actual implementations may be described in the specification. It should be understood that, as in any engineering or design project, the development of any such actual implementation requires numerous implementation-specific decisions to achieve the developer's specific objectives, such as complying with system-related and business-related constraints that may differ between implementations. Furthermore, it should be understood that such development efforts may be complex and time-consuming, but remain routine tasks of design, fabrication, and manufacturing for those skilled in the art who benefit from this disclosure.

[0036] When describing elements of various embodiments of the invention, the articles “a,” “an,” “the,” and “the” are intended to indicate the presence of one or more such elements. The terms “comprising,” “including,” and “having” are intended to be inclusive and indicate that additional elements may exist in addition to those listed. Furthermore, any numerical examples in the following discussion are intended to be non-limiting, and therefore any additional numerical values, ranges, and percentages are within the scope of the disclosed embodiments.

[0037] As used herein, “electrical coupling,” “electrical connection,” and “electrical communication” mean that the referenced components are directly or indirectly connected so that current can flow from one to the other. This connection may include direct conductive connections (i.e., without intervening capacitors, inductors, or active components), inductive connections, capacitive connections, and / or any other suitable electrical connection. Intervening components may be present.

[0038] Furthermore, although the embodiments disclosed herein are described with respect to a mammography apparatus for digital breast tomography (DBT), it should be understood that embodiments of the invention are applicable to other types of imaging devices used for both two-dimensional and three-dimensional imaging (including, for example, fluorescence fluoroscopy, full-field digital mammography, two-dimensional imaging of breast tissue, and energy dispersive mammography (single-energy or multi-energy)), as well as other types of imaging devices used for imaging procedures of tissues other than breast tissue. Moreover, embodiments of the invention are generally applicable to tissue analysis, but are not limited to the analysis of human tissue.

[0039] See now Figure 2 and Figure 3 This illustration shows the main components of an exemplary imaging system 10, configured as a mammography system 12 for imaging breast tissue according to an embodiment of the present invention. System 10 includes a radiation source / X-ray source 16, a radiation detector 18, and a collimator 20, the system being described in various names, such as apparatus and method for mammography involving breast compression. Apparatus And Method For Mammographic Breast Compression The system disclosed in U.S. Patent Application Publication No. US20200060632, the entire contents of which are expressly incorporated herein by reference for all purposes. Radiation source 16 is movable relative to detector 18 between multiple imaging positions and operable to emit radiation rays 22 ( Figure 3 These radiation rays are received by radiation detector 18 to provide an image of an object such as breast 52. In an embodiment, system 10 may include a patient shield 24 mounted to radiation source 16 via a face shield rail 26 to prevent the patient's head from obstructing the radiation rays and to protect the patient from radiation ray 22.

[0040] Further reference is still needed. Figure 2 and Figure 3 System 10 also includes a motorized and / or manually adjustable pressure paddle or plate 28 and a support structure 30, to which one or more of the radiation source 16, radiation detector 18, and / or pressure plate 28 are mounted. In one embodiment, system 10 includes a controller 32. The controller 32 may be a workstation having at least one processor / central processing unit / computer and a memory device / database storing information and / or non-transitory instructions adopted by the controller 32 for various operating modes of system 10, such as... Figure 1As shown, or in other embodiments, controller 32 may be embedded / integrated into one or more of the various components of the system 10 disclosed above. In embodiments, controller 32 may be electrically connected to radiation source 16, radiation detector 18, and / or pressure plate 28 via cable 34. As will be understood, in embodiments, connection 34 may be a wireless connection. In embodiments, controller 32 may include a radiation shield 36 that protects the operator of system 10 from radiation rays 22 emitted by radiation source 16. Controller 32 may also include a display 38, keyboard 40, mouse 42, and / or other suitable user input devices that facilitate control of system 10 via user interface 44.

[0041] like Figure 2 and Figure 3 As further shown, radiation source 16, together with radiation detector 18, forms part of an X-ray system that provides X-ray images for the purpose of imaging a body part of a patient, such as breast 52. As described above, radiation source 16 emits radiation rays 22 such that the radiation rays 22 travel from radiation source 16 to radiation detector 18. Although radiation rays 22 are discussed herein as X-rays, it should be understood that radiation source 16 may emit other types of electromagnetic rays that can be used for imaging a patient. Radiation source 16 may be mounted to support structure 30 such that the radiation source can rotate relative to radiation detector 18 about axis 46, but movement of radiation source 16 in paths other than rotation about a fixed axis is also contemplated, such as during digital breast tomography (DBT). In embodiments, radiation detector 18 may be configured to rotate or translate within its housing, such as in the directions indicated by arrows 53 and 55.

[0042] exist Figure 2 In the illustrated exemplary embodiment, radiation source 16 and detector 18 are mounted to a rack 90 fastened to a support structure 30. The support structure 30 houses a translation mechanism 92 operatively connected to the rack 90. ​​The translation mechanism 92 is operable to vertically move the rack 90 relative to the support structure 30 to position the rack 90 at an appropriate height to accommodate the size of a patient utilizing the system 10 thereon. The translation mechanism 92 is also operable to rotate the rack 90 about a horizontal axis 46 relative to the support structure 30 to rotatably position the rack 90 relative to the patient as needed.

[0043] The gantry 90 includes a generally C-shaped body 94, with a radiation source 16 at one end and a detector 18 at the opposite end. In this configuration, regardless of the vertical and / or rotational orientation of the gantry 90, such as positioning the radiation source 16 and detector 18 relative to the patient's breast 52 to obtain X-ray images in various orientations, such as for a head-to-tail (CC) or median-lateral oblique (MLO) view, etc., the radiation source 16 is aligned with the detector 18. In this position, the detector 18 is able to receive X-rays 22 emitted from the radiation source 16, which pass through the patient portion, i.e., the patient's breast 52, located between the radiation source 16 and the detector 18, to generate image data to be transmitted to the control system 32 of the mammography equipment / system 10 to create a 3D image dataset for physician review (such as by using DBT), and other known methods.

[0044] Alternatively, in another embodiment, the radiation source 16 may be attached to the gantry 90 to rotate and / or move independently of the gantry 90 and detector 18, so that the radiation source 16 can take multiple X-ray projections / images of the patient's breast at various locations within a limited angular range relative to the detector 18, for example, at angles between + / -7.5°, + / -12.5°, + / -15°, + / -25°, and + / -60° relative to the object and / or the vertical axis 100. Images obtained by the radiation source 16 at multiple locations between these angles can be used to create stereoscopic images in biopsy procedures using system 10 or for DBT when operating system 10 in imaging mode.

[0045] As described above, radiation detector 18 receives radiation rays 22 emitted by radiation source 16. In an embodiment, data regarding the radiation rays 22 received by radiation detector 18 can be electrically transmitted from radiation detector 18 to controller 32 via cable / electronic connection 34, such that controller 32 generates one or more images that can be displayed on display 38 and stored in a memory device.

[0046] The compression plate 28 and the motor (not shown) controlling its movement are operable in response to commands from controller 32 or from a controller such as remote control 84 on or near the mammography system 10 or a switch controller 80 connected via cable 82, to move toward and away from the radiation detector 18, as indicated by arrow / compression axis 48, such that the compression plate 28 flattens and holds a body part (e.g., breast 52) ​​against the surface 50 of the radiation detector 18 in place. In this respect, the radiation detector 18 and its surface 50 are referred to herein as the “compression surface or support plate,” which cooperates with the compression plate 28 to compress and clamp the patient’s breast between them.

[0047] In one embodiment, system 10 may further include or alternatively include Figure 3 The biopsy tool 120 is illustrated in the diagram. In this embodiment, the radiation source 16, together with the radiation detector 18, forms part of an X-ray system that provides X-ray images for guiding the biopsy tool 120 (e.g., a needle) to a suspicious site within the patient's body. Figure 3 As shown, in the embodiment, the biopsy tool 120 may be mounted on the biopsy locator 110 mounted to the support structure 30, such that it also rotates about axis 46 in a manner similar to radiation source 16 and / or moves in the vertical and / or horizontal directions in a manner similar to pressure plate 28.

[0048] Now for reference Figure 4 An exemplary embodiment of a method 400 for operating a mammography system 10 is illustrated. In method 400, in an initial step 402, a radiation source 16 is operated to obtain multiple projected images 404 of the object / breast 52 at multiple different angular positions of the radiation source 16 relative to the breast 52 and the detector 18, for example, within a limited angular range (e.g., less than 180°), where a computed tomography (CT) imaging procedure requires at least 180° of angular range, or less than + / - 60°, or less than + / - 25°, or less than + / - 7.5°, each angular range relative to a vertical axis 100 such as that used in a DBT imaging procedure. Figure 3 In step 406, these projected images 404 are used by a conventional reconstruction algorithm 408 to form a standard reconstruction volume 410, which can be stored as a set of processor-executable and / or non-transitory instructions in memory within the system 10 accessible by the processor 32.

[0049] In step 412, the standard reconstruction volume 410 is utilized by one or more individual artifact reduction and / or resolution enhancement algorithms / artificial intelligence (AI) / neural network (NN) 414 to form an artifact-reduced or quasi-isotropic volume 416. Examples of one or more artifact reduction and / or resolution enhancement algorithms / AI / NN 414 designed for DBT can be found in the following literature: D. Wu, K. Kim and Q. Li, “Digital Breast Tomosynthesis Reconstruction with DeepNeural Network for Improved Contrast and In-Depth Resolution”, in: 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI), Apr. 2020, pp.656–659; A. Quillent et al., "A Deep Learning Method Trained on Synthetic Data for Digital Breast Tomosynthesis Reconstruction", Medical Imaging with Deep Learning , 2023; and A. Quillent et al., “Deep-Learning Uncertainty Estimation for Data-Consistent Breast Tomosynthesis Reconstruction”, in: 21st International Symposium on Biomedical Imaging (ISBI 2024), IEEE Signal Processing Society and IEEE Engineering in Medicine and Biology Society, Athens, Greece, 2024; each of these references is explicitly incorporated herein by reference in its entirety for all purposes. The resolution enhancement algorithm / AI / NN 414 reduces artifacts included in the standard reconstruction volume 410, particularly those exemplified within images along planes not parallel to the plane of detector 18 (e.g., planes orthogonal to detector 18, such as the coronal and sagittal planes), thereby significantly improving the image quality of those images and enabling them to be efficiently read for diagnostic and other purposes. In addition to the artifact reduction volume 416, the resolution enhancement algorithm / AI / NN 414 also outputs an uncertainty map 418, which graphically illustrates the evaluation of the quality of the artifact reduction or quasi-isotropic volume 416 and any image generated therefrom. Regions where the resolution enhancement algorithm / AI / NN 414 performs poorly are shown as high pixel intensities in the uncertainty map 418.

[0050] In an alternative implementation, the processing of the projected image 404 by the first reconstruction algorithm 408 is omitted, and the resolution enhancement algorithm / AI / NN 414 can be directly applied to the projection 404 (i.e., typically in the attenuation domain), producing an artifact reduction volume 416. Examples of artifact reduction and / or resolution enhancement algorithms / AI / NN 414 directly applied to the projected image 404 with the first reconstruction algorithm stage 408 omitted can be found in the following literature: J. Forest et al., “Deep learning reconstruction of digital breast tomosynthesis images for accurate breastdensity and patient-specific radiation dose estimation”. Medical Image Analysis , vol. 71, p. 102061, Jul. 2021, which is hereby explicitly incorporated in its entirety by reference for all purposes.

[0051] Referring to the steps performed in method 400 and the various post-processing steps to be described, the operation of each of the reconstruction algorithm 410 and the resolution enhancement algorithm / AI / NN 414 can be executed by processor 32 using processor-executable and / or non-transitory instructions stored in a memory device and accessible by processor 32, relating to the operation of the mammography system 10, the reconstruction algorithm 410, and the resolution enhancement algorithm / AI / NN 414.

[0052] Looking at it now Figures 5 to 1 5. Once the artifact reduction volume 416 has been generated as the output from the resolution enhancement algorithm / AI / NN 414 in step 412, different post-processing steps can be performed using the artifact reduction volume 416.

[0053] Special reference Figures 5 to 6 With the aid of artifact reduction volume 416, a view or image of the breast 52 can be generated at quasi-isotropic resolution along previously unused viewing directions or planes within the artifact reduction and / or quasi-isotropic volume 416 (such as along a plane not parallel to detector 18, along a plane orthogonal to detector 18, or along any line between detector 18 and the focal point of radiation source 16). Figure 5The image illustrates a display 38 of a mammography system 10, showing one or more standard reconstructed images, such as a cross-sectional image 419 of the breast 52 formed by a standard reconstruction volume 410. Furthermore, coronal images 422 and sagittal images 424, formed by artifact reduction and / or quasi-isotropic volumes 416, are presented in conjunction with the cross-sectional image 419. In some embodiments, additional image planes, such as the coronal image 422 and sagittal image 424, each have at least a quasi-isotropic resolution similar to that of the cross-sectional image 419 obtained from the standard reconstruction volume 410, such that each provides useful diagnostic information about the imaged breast 52, which is consistent with the imaged breast 52. Figure 1 The images shown are generated only from these identical planes in contrast. Referring to the artifact-reducing and / or quasi-isotropic volume 416 and the enhanced quasi-isotropic resolution provided within various planar images (e.g., the cross-sectional image 420, coronal image 422, and / or sagittal image 424 from which it is generated), the operation of the resolution enhancement algorithm / AI / NN 414 on the standard reconstruction volume 410 provides the artifact-reducing and / or quasi-isotropic volume 416, which includes edge information within or near the full 180° range of the breast 52, representing typical breast texture, such as… Figures 18A to 18B The illustrations in the text, as well as examples such as Quinto, ET, Artifacts and Visible Singularities in Limited Data X-Ray Tomography, Sens Imaging As described in , 9 (2017), the entire contents of that document are expressly incorporated herein by reference for all purposes.

[0054] At the current level of technology, the resolution in a plane of conventional DBT is typically tens of millimeters or better, meaning the point spread function in the plane is typically 250 mm wide (FWHM). Furthermore, the current resolution of conventional DBT in the lateral direction is highly dependent on the size of the object in the planar direction. Therefore, the artifact spread function associated with this plane can be calculated based on the object size and scan angle, resulting in a resolution of centimeters or worse. (See Dalmonte et al., DOI: 10.1016 / j.ejmp.2024.103300, which is incorporated herein by reference in its entirety for all purposes). Resolution enhancement algorithm / AI / NN 414 provides a quasi-isotropic improvement in resolution in image planes 422, 424 of a quasi-isotropic volume 416, which are orthogonal to detector 18, not parallel to detector 18, and / or oriented from centimeters to millimeters in the z-direction along any line (e.g., coronal and sagittal planes) between detector 18 and the focal point of radiation source 16.

[0055] also, Figure 6 Transverse, coronal, and sagittal simulated images of the breast 52, compressed and positioned as in a standard DBT system, are presented, as obtained from a volume generated in a computed tomography (CT) imaging procedure that does not produce the same artifacts in a plane not parallel to detector 18. As shown, the coronal image 422 and sagittal image 424 obtained from artifact reduction and / or quasi-isotropic volume 416 are very close along the path from the CT imaging procedure ( Figure 6 The isotropic resolution of the images obtained in the same plane enables verification of the accuracy of the image data in the artifact reduction and / or quasi-isotropic volume 416, as well as various images (e.g., the coronal image 422 and sagittal image 424 generated therefrom, or any other image along a plane not parallel to detector 18).

[0056] exist Figure 7 In an alternative embodiment shown, the cross-sectional image 420 presented on display 38 may also be generated from the artifact reduction and / or quasi-isotropic volume 416. Furthermore, as an alternative and / or supplement to one or more of the cross-sectional image 420, coronal image 422, and sagittal image 424, system 10 and / or processor 32 may utilize some specific post-processing to reproject one or more images along any unacquired angle within the artifact reduction and / or quasi-isotropic volume 416, such as, but not limited to, generating synthetic 2D images in any direction within the artifact reduction and / or quasi-isotropic volume 416, including along each of the coronal / cross-sectional / sagittal planes and other planes depending on the application, and not limited to images along the plane defined by the angle of the radiation source 16 that acquired the projected image 404.

[0057] See now Figure 8 A, as Figure 5 The coronal image 422 and sagittal image 424 shown in the figure and / or Figure 7 Instead of the cross-sectional image 420, to reduce the amount of data to be read for the user or transmitted in step 425, the processor 32 is operable to form the artifact reduction and / or quasi-isotropic volume 416 as a group or sheet 426 comprising planes including similar image data. The sheet 426 may be oriented along any of the cross-sectional, coronal, and / or sagittal planes, resulting in enhanced image quality of the artifact reduction and / or quasi-isotropic volume 416. The thickness of the sheet 426 on the artifact reduction and / or quasi-isotropic volume 416 may be determined by the processor 32 based on analysis of the similarity of data on the planes of the artifact reduction and / or quasi-isotropic volume 416 or based on a manually selected thickness for the sheet 426.

[0058] Or, such as Figure 8 As shown in the exemplary embodiment of B, in step 427, processor 32 is operable to generate a synthetic 2D image 428 along each of the transverse, coronal, and / or sagittal planes and / or any other desired plane not parallel to detector 18, resulting in artifact reduction and / or image quality enhancement of the quasi-isotropic volume 416. The synthetic 2D image 428 effectively summarizes information across the entire artifact reduction volume 416 in selected planes (transverse, coronal, sagittal) of a single 2D image to provide an image capable of rapid viewing and easy evaluation of all image data within the artifact reduction and / or quasi-isotropic volume 416.

[0059] In addition, Figure 7 In this embodiment, display 38 is illustrated as presenting multiple images 420, 422, 424 along the transverse, coronal, and sagittal planes, as well as other desired planes not parallel to detector 18, as discussed above, at least some of which are obtained from artifact reduction and / or quasi-isotropic volume 416. Alternatively to images 420, 422, 424, display 38 may present images generated along different directions or planes using different slicing algorithms. For example, artifact reduction volume 416 may be presented as a single 3D slice volume shown from different directions (e.g., transverse, coronal, and sagittal). Alternatively, display 38 may present multiple 2D slice volumes 430, 432, 434 synchronized on display 38 (e.g., 1 for each of the transverse, coronal, and sagittal planes). Furthermore, similar to... Figure 5 In images 420, 422, and 424, the cross-sectional slice volume 426 can be formed from the standard reconstruction volume 410, while the coronal slice volume 428 and sagittal slice volume 430 can be formed from the artifact reduction volume 416. The thickness of slices 430, 432, and 434 generated at different orientations (e.g., cross-section and coronal) can be different.

[0060] Now for reference Figure 9 In addition to or separate from the images 420-424 and the slice volumes 430-434 described above, an enhanced volumetric rendering 436 can also be formed from the artifact reduction and / or quasi-isotropic volume 416. The enhanced volumetric rendering 436 is a representation of the artifact reduction and / or quasi-isotropic volume 416, which makes it possible to view the entire external and internal structure of the imaged breast 52 at various rotational and cross-sectional positions of the enhanced volume 432.

[0061] Looking at it now Figure 10 In a specific configuration, the mammography system 10 includes a biopsy locator 110. Figure 2 ) and the biopsy device 120 positioned on it ( Figure 2This biopsy device can be operated by a user via control system 32 to perform biopsy procedures. In existing technology systems, such as Figure 10 As shown, during the biopsy procedure, a simplified breast map 500, identical for all patients, is displayed on monitor 38 as a map of the movement of the guide needle 510 toward the biopsy target 504. During the execution of the biopsy procedure, the target 504 is selected on an in-plane image 506 (similar to 419) of the standard reconstruction volume 410. The expected needle trajectory 502 is also displayed on graph 500 along with a calculated safety margin 508 for trajectory 502. Trajectory 502 cannot be shown on the standard reconstruction planes 506 and 419 because the image quality is too low. The needle 510 is then inserted into the breast 52, and the clinician can follow the progress of the needle 510 on graph 500.

[0062] See now Figure 11 The improved image quality of the artifact reduction and / or quasi-isotropic volume 416 generated by the resolution enhancement algorithm / AI / NN 414 allows for the replacement of graph 500 with artifact reduction and / or quasi-isotropic volumes 416, 432, or images 420, 422, 424. Therefore, the display 38 presents a personalized map of the needle trajectory 502, target 504, and safety margin 508 for each patient. Using artifact reduction and / or quasi-isotropic volumes 416, 432, or images 420, 422, 424 as the map on the display 38 also allows for more precise illustration of the needle 510's position on the map and easier visualization and determination of the size and location of the target 504.

[0063] Looking at it now Figure 12 , Figure 4 The outputs of the different steps 406 and 412 of method 400 include a standard reconstruction volume 410 and an artifact-reduced and / or quasi-isotropic volume 416. Since these corresponding volumes 410, 416 are formed from the same projected image 404 of the breast 52 under a single compression, the volumes 410, 416 can be registered with each other by system 10 and / or processor 32 in any suitable known manner. Due to this registration or synchronization between the standard reconstruction volume 410 and the artifact-reduced and / or quasi-isotropic volume 416 and / or quasi-isotropic images 420, 422, 424, the user can use the user interface 44 to select the focal point or region on the standard reconstruction volume 410 presented on the display 38, or on the focal point or region on the standard reconstruction image / virtual image / projection created from it, and the processor 32 can present one or more orthogonal views, for example, along the cross section, coronal plane and / or sagittal plane, or other virtual projections generated from the artifact-reduced and / or quasi-isotropic volume 416 at the same registration point in the artifact-reduced and / or quasi-isotropic volume 416.

[0064] By utilizing the standard reconstruction volume 410 / standard reconstruction image 419 and the artifact reduction and / or quasi-isotropic volume 416 / quasi-isotropic images 420, 422, 424 for registration and / or synchronization, when images 419, 420, 422, 424 are such as in Figure 5 When presented on the display 38 as described above, the movement of cursor 550 by the user on one of the images 419, 422, 424 can be synchronized with the movement of individual cursor 550' set on one or more of the other images 419, 422, 424, thereby illustrating views along different planes or axes to show the position of cursors 550, 550' in each of the displayed images 419, 422, 424.

[0065] Furthermore, due to the registration / synchronization between volumes 410 and 416 and the associated reconstructed and quasi-isotropic images, a user interface 44 can be used to manually annotate a three-dimensional region of interest (ROI) 600 in an image 419, 420, 422, 424 or volumes 426, 428, 430, 432, and to instruct and calculate the volume and diameter of the same ROI 600 in all other virtual projections, images, volumes, and / or axes, in contrast to the limited two-dimensional annotation available on a single image or axis on current mammography systems. Figure 12 As shown, the annotation 602 of a single ROI 600 in the cross-sectional images 419, 420, and 604 of the breast 52 can be converted into a volume 606 represented in the separately displayed coronal images 422 and 608 of the breast 52.

[0066] Annotation 602 can also be utilized by processor 32 to perform automatic or semi-automatic calculation of the volume of the mass / lesion / ROI 600 identified within annotation 602, which is currently not feasible in DBT. The result can be stored and accessed later and / or used to track any changes in the volume of the mass / lesion / ROI 600 over time, and optionally to propose a projection of expected progression based on the form and / or composition of the volume of the mass / lesion / ROI 600.

[0067] See now Figure 13 With the aid of the uncertainty graph 418 generated together with the artifact reduction and / or quasi-isotropic volume 416, an indication 700 of the calculated uncertainty of the virtual projections or images 420, 422, 424 and / or volumes 426, 428, 430 generated from the artifact reduction and / or quasi-isotropic volume 416 can be added. Figure 13In the exemplary embodiment illustrated, indication 700 may take the form of a color overlay 702 located over the displayed images 420, 422, 424 and / or volumes 426, 428, 430, wherein different colors represent various levels of uncertainty within the individual pixels or voxels of the images 420, 422, 424 and / or volumes 426, 428, 430 generated from the artifact reduction volume 416. In addition to or as an alternative to indication 700, the displayed images 420, 422, 424 and / or volumes 426, 428, 430 may include a total uncertainty score 704, which represents the overall computational uncertainty of the images 420, 422, 424 and / or volumes 426, 428, 430 generated from the artifact reduction and / or quasi-isotropic volume 416. In addition to indication 700, if the uncertainty score of the displayed images 420, 422, 424 and / or volumes 426, 428, 430 is higher than a predetermined threshold (which can be preset automatically or manually), the system 10 and / or processor 32 may present a warning to the user on the display 38 regarding the high uncertainty value.

[0068] Looking at it now Figures 14A to 14D After outputting the artifact reduction and / or quasi-isotropic volume 416 from method 400, conventional image segmentation can be performed on the artifact reduction and / or quasi-isotropic volume 416 to assign material (e.g., glands, fat, masses / cysts, etc.) to each voxel in the artifact reduction and / or quasi-isotropic volume 416 using standard image segmentation algorithms currently used in magnetic resonance imaging (MRI) and CT imaging. Examples of these procedures are disclosed in each of the following references: Frangi, AF, Prince, JL, and Sonka, M., (2024). Medical Image Analysis Elsevier (DOI: https: / / doi.org / 10.1016 / C2015-0-06316-X), particularly regarding the conventional methods described in Part 3 and the AI ​​methods described in Chapter 18 of Part 5; and / or N. Geeraert et al., 2014 Phys.Med.Biol. 59 4391; Comparison Of Volumetric Breast Density Estimations From Mammography And Thorax CT (DOI: 10.1088 / 0031-9155 / 59 / 15 / 4391), the full contents of which are hereby expressly incorporated in this paper by reference for all purposes.

[0069] Knowing the physical properties of each material type assigned to each voxel in the artifact reduction and / or quasi-isotropic volume 416 via image segmentation of the artifact reduction and / or quasi-isotropic volume 416, and the compressive forces applied to the breast 52 during the imaging procedure executed by the system 10 (as recorded by the system 10 via the compression paddle 28), the processor 32 can numerically invert the compression and gravity represented in the artifact reduction and / or quasi-isotropic volume 416 to create, for example, Figures 14A to 14B The figure of uncompressed breast volume 800 is schematically illustrated. The figure of uncompressed volume 800 represents and can be shown as the form of breast 52 under the influence of gravity alone before any compressive force is applied to it.

[0070] like Figures 14C to 14D As shown, the uncompressed digital volume 800 can be digitally compressed to form a first modified digital volume 800', which is numerically modified / digitally compressed to show the form of the breast 52 under the action of a simulated first compression force applied to the breast 52 by the compression surface 50 of the compression paddle 28 and the detector 18. Additionally, in Figures 14C to 14D In the second modified digital volume 800", it illustrates another digital compression, which shows the effect of a simulated second compression force applied to the breast 52 by the compression paddle 28 and the compression surface 50 of the detector 18.

[0071] like Figures 14A to 14D As shown, the digital uncompressed breast volume 800 created from the artifact reduction and / or quasi-isotropic volume 416 can be oriented to any simulated location and digitally compressed in any desired direction with a specified force to generate a synthetic image of the breast 52. The optimal compression direction, compression paddle position, and breast position on the compression surface can be selected according to specific diagnostic needs for digital manipulation of the digital uncompressed breast volume 800. Therefore, the digital uncompressed breast volume 800 can be used to create simulated volumes (800', 800''') from which virtual or synthetic images of the breast 52 can be obtained, such as synthetic cephalothorax (CC) and / or medial-lateral oblique (MLO) images of the breast 52 for diagnostic purposes.

[0072] Furthermore, when system 10 and / or processor 32 are capable of generating multiple synthetic images of breast 52, where each image created from the digital uncompressed breast volume 800 is effectively registered with each other, when multiple synthetic images of breast 52 (such as synthetic CC images and synthetic MLO images) are displayed together, the user can use user interface 44 to select objects and / or regions within the CC image, and the corresponding objects or regions can be shown in the MLO image in a much more accurate manner than currently possible. Alternatively, a separate method 400 can be performed on breast 52 in each of the CC and MLO orientations, where a digital uncompressed breast volume 800 is formed for each orientation. The resulting CC digital uncompressed breast volume and MLO digital uncompressed breast volume can be registered with each other such that selection of an object within one of the CC digital uncompressed breast volume or the MLO digital uncompressed breast volume will exemplify the same object in the image or plane of the other of the CC digital uncompressed breast volume and the MLO digital uncompressed breast volume.

[0073] Furthermore, synthetic images (e.g., synthetic CC and / or MLO images) can be combined or used in conjunction with actual images of the breast 52 (e.g., actual CC and / or MLO images) to enhance the actual image of the breast 52.

[0074] Looking at it now Figures 15A to 15C Artifact reduction and / or quasi-isotropic volume 416 can also be used to assess the tissue forming the breast 52. More specifically, to determine the density of the breast 52, as previously referred to... Figures 14A to 14D The image segmentation algorithm can be used to process the artifact reduction and / or quasi-isotropic volume 416 to allocate material to each voxel in the artifact reduction volume 416. By utilizing the results of image segmentation, for example, assigning material or tissue type to specific voxels in the artifact reduction and / or quasi-isotropic volume 416 based on the intensity values ​​of voxels within the artifact reduction and / or quasi-isotropic volume 416, breast density can be calculated as the ratio of the number of voxels classified as glandular tissue to the number of voxels classified as fat.

[0075] Specifically, in Figure 15A In an exemplary implementation, an image of breast 52 obtained from a standard reconstruction volume 410 is shown for comparison with images obtained from... Figure 15B The images obtained by artifact reduction and / or quasi-isotropic volume 416 are compared, where Figures 15A to 15B The same plan view of breast 52 is presented. By utilizing Figure 15B The images in the image can be obtained using the results of the previously described image segmentation (e.g., for analysis). Figure 15B To create (the intensity of each pixel in the image) Figure 15CMaterial diagram / image 900, which illustrates the types and locations of various tissues presented in the image. (Assigned to...) Figure 15C In images where the illustrated tissue category of pixels is "glandular" or "fatty," breast density can be calculated as the ratio of the number of pixels classified as glandular to the number of pixels classified as fatty. Therefore, Figure 900, formed by system 10 and / or processor 32, provides an enhancement to the utility of system 10 for quantitative imaging of the breast, where system 10 and / or processor 32 can define and display new units, such as the calibrated glandular / fatty ratio of the imaged breast 52, similar to the Henle units used in CT imaging, to provide a better estimate of glandular density and location, thereby enabling a more accurate calculation of the radiation dose received by the patient and / or the average glandular dose.

[0076] See now Figure 16 By leveraging the ability of artifact reduction and / or the quasi-isotropic volume 416 to provide images in a plane not perpendicular to detector 18, system 10 and / or processor 32 can additionally perform specific mechanical imaging procedures on breast 52, including compression or strain X-ray elastography. Since cancerous and non-cancerous lesions, as well as other tissues of different types (such as other benign and malignant lesions), will exhibit different amounts of tissue movement relative to the normal surrounding breast tissue when minimal pressure is applied, strain X-ray elastography is a qualitative method based on measuring stiffness by the soft tissue deformation and / or displacement caused by different levels of pressure applied to the tissue.

[0077] like Figure 16 As illustrated schematically, in the operation of the mammography imaging system 10, to perform X-ray imaging of the breast 52 for screening or diagnostic purposes, a pressure plate or pressure paddle 28 with a specified or selected pressure force (such as via a motor operatively connected to the pressure plate 28 and controlled by the controller 32) is used to press the breast 52 against and against the detector 18. The pressure force applied by the pressure plate 28 is monitored and recorded by the processor 32, so that the amount of pressure force applied to the breast 52 can be precisely controlled.

[0078] As previously stated, to provide useful diagnostic information about the imaged breast 52, the prior art image of the breast 52 obtained using the mammography imaging system 10 needs to be either a projected image 404 perpendicular to the orientation of the detector 18 and the compression plate 28 (which has significantly low resolution and is therefore useless for diagnostic purposes) or a reconstructed planar image parallel to the orientation of the detector 18 and the compression plate 28 (which has much higher resolution and can be used for diagnostic purposes). Therefore, prior art reconstructed images along the cross-section are "blind" to displacement in the direction of compression (e.g., changes in the height of any lesion within the breast 52). By enhancing the image obtained by the mammography imaging system 10... Figure 4The method provides a high resolution for reconstructed images oriented along a plane orthogonal to detector 18 and compression plate 28 (i.e., in the direction of the compressive force and the resulting displacement), which can adapt elastography to breast tomography fusion.

[0079] More specifically, see reference Figure 2 , Figure 4 , Figure 16 and Figure 17 In method 1000, firstly, in step 1002, the breast 52 is positioned on the compression surface 50 of the detector 18, and the compression paddle 28 is moved to compress the breast 52 under normal or first compression force. In step 1004, a DBT imaging procedure is performed on the breast 52 to obtain a first set of projected images 404 of the breast 52 at various angular positions of the radiation source 16 relative to the detector 18. In step 1006, in... Figure 4 In method 400, a first set of projected images 404 is used to generate a first artifact reduction and / or quasi-isotropic volume 416. In step 1008, the pressure paddle 28 moves vertically to generate small changes (increase or decrease) in the pressure force on the breast 52, and / or moves horizontally toward or away from the gantry 90 to generate differential displacement of structures and / or lesions within the breast 52 due to the hysteresis of the lesion-forming tissue, which, depending on the type of lesion-forming tissue, at least partially exhibits an elastic or inelastic mode. In a subsequent step 1010, a second DBT imaging procedure is performed on the breast 52 to obtain a second set of projected images 404' of the breast 52 at various angular positions of the radiation source 16 relative to the detector 18. In step 1012, in Figure 4 In method 400, a second set of projected images 404' is used to generate a second artifact reduction and / or quasi-isotropic volume 416'.

[0080] In step 1014, images 1018 and 1020 of similar planes orthogonal to the plane of detector 18 are obtained from each artifact reduction and / or quasi-isotropic volume 416, 416', and are jointly utilized or compared by processor 32 and / or other external computing devices (not shown) to provide information or evidence of any lesions present in images 1018 and 1020, differential local displacement and / or mechanical properties, to diagnostically determine the form of those lesions using any one or more known methods (such as subtraction, identification of similar structures, determination and display of displacement vectors or their modules between source points as images, correlation images, etc.). Some examples of the analytical process to be employed may be any one or more of the following methods, each of which is expressly incorporated herein by reference in its entirety for all purposes: Ramião, NG, Martins, PS, Rynkevic, R. et al. ,Biomechanical properties of breast tissue, a state-of-the-art review, Biomech Model Mechanobiol 15, 1307–1323 (2016); Zhou, Cong, Hainsworth, Brent, Sydney, Maxwell, Lee, Michael, Ormsby, Zane, Haggers, Marcus and Chase, J. Geoffrey, “Structural health monitoring of tissue mechanics for non-invasive diagnosis of breast cancer” at - Automatisierungstechnik , vol. 66, no. 12, 2018, pp. 1037-1050; Nitta, N., Shiina, T. (2005), BreastTissue Assessments Based on High Order Mechanical Properties, in: Ueno, E., Shiina, T., Kubota, M., Sawai, K. (eds) Research and Development in BreastUltrasound, Springer, Tokyo.

[0081] Using this information, artifact reduction and / or quasi-isotropic volumes 416, 416' can be used to assess the malignancy of any lesion within the imaged breast 52 based on the determination of the stiffness of the tissue forming the lesion according to images 1018, 1020 of similar planes orthogonal to the plane of detector 18 generated from artifact reduction and / or quasi-isotropic volumes 416, 416'.

[0082] Furthermore, artifact reduction and / or quasi-isotropic volume 416 can be used in conjunction with images and / or volumes of the breast 52 obtained using other imaging modalities (including but not limited to MRI, bCT, and ultrasound (US)) to enhance its diagnostic utility. For example, as referenced Figure 12As discussed, due to the correspondence between the geometry in the corresponding views / images 420, 422, 424 and the artifact reduction volume 416 from which the views / images 420, 422, 424 are generated, annotations made in views, sections, or images along one plane 420, 422, 424 can be translated into views or images along another plane 420, 422, 424. Furthermore, if the volume of the breast 52 is obtained using a different imaging modality (including but not limited to US or MRI), this volume can be registered with the artifact reduction and / or quasi-isotropic volume 416 using known cross-modal registration procedures. Therefore, annotations 602 made in views / images 420, 422, 424 can be translated into alternative modal volumes, and vice versa. Therefore, annotation 602 made in one of the artifact reduction and / or quasi-isotropic volume 416 or alternative modal volume can be adopted by system 10 and / or processor 32 to determine and present selected views or images from the alternative modal volume and images 420, 422, 424 along each plane (i.e., transverse, coronal, sagittal) that intersect at annotation 602 to achieve precise localization of lesions or tumors across modalities.

[0083] In yet another exemplary embodiment, the pressure paddle 28 can be formed of a material transparent to ultrasound, such as polymethylpentene sold by Mitsui Chemicals of Japan under the name TPX. First, the pressure paddle 28 can be used to compress the breast 52 to perform the DBT imaging procedure / acquisition as described herein on the compressed breast 52. Subsequently, an artifact-reduced / quasi-isotropic DBT volume 416 can be generated based on the projection image 404 obtained from the acquisition. With the aid of the pressure paddle 28 made of this type of material, after compression of the breast 52 and the performance of the DBT imaging procedure / acquisition on the compressed breast 52, an ultrasound imaging system / ultrasound probe (not shown) can be used to obtain ultrasound (U / S) images through the paddle while keeping the breast 52 under the same pressure as during the DBT acquisition. With the aid of artifact reduction and / or quasi-isotropic volume 416, the plane of the U / S image obtained through breast 52 can be identified, for example, the plane orthogonal to the compression paddle 28 on which the U / S probe is positioned. After calibration is performed between the artifact reduction and / or quasi-isotropic volume 416 and the U / S image and / or the system by using depth, by manually aligning the identified result in the U / S image with the image plane of the artifact reduction and / or quasi-isotropic volume 416, and / or by using known characteristics of the U / S system (e.g., probe type, U / S frequency, and characteristics of breast tissue, etc.), the U / S image is superimposed on or over the quasi-isotropic volume 416.

[0084] Furthermore, with the aid of this configuration for the imaging system 10 and the U / S transparent compression paddle 28, a projected image 404 of the compressed breast 52 can be obtained to generate the artifact reduction and / or quasi-isotropic volume 416 as described above. The user and / or a computer system such as the processor 32 can then identify any suspicious results at a particular location on the breast 52, for example, within a conventional or standard image plane or within a plane obtained from the quasi-isotropic volume 416 according to the previously described method.

[0085] If a suspicious result is identified, breast 52 can be further analyzed by obtaining a U / S image of breast 52 in a plane benefiting from the invention (i.e., clearly perpendicular to paddle 28), wherein the U / S image can be acquired under the same compression or with similar compression after repositioning breast 52 to a clearly similar location. After obtaining the U / S image, it can be registered with the quasi-isotropic / artifact reduction volume 416 and / or any DBT plane image formed therefrom, according to the methods of this disclosure previously discussed, in order to enhance the diagnostic capability of imaging system 10 in a different and improved manner than simple re-registration of images obtained from different imaging modalities.

[0086] Furthermore, regarding the reduction of volume from artifacts by 416 to generate a digital uncompressed breast volume of 800, as previously referenced... Figures 14A to 14D The digital uncompressed breast volume 800 or artifact-reduced volume 416 discussed can be visually compared with images and / or volumes of the breast 52 obtained using other imaging modalities.

[0087] For example, the digital compression volume of the breast 52 obtained from a US or MRI imaging procedure can be compared with that obtained via... Figure 4 The method obtained by the method was compared with the reduction in breast volume of 416 by 52 artifacts.

[0088] In addition, the uncompressed volume of the breast 52 obtained from a US or MRI imaging procedure can be compared with the digital uncompressed volume 800 generated from the artifact reduction volume 416.

[0089] Furthermore, the digital uncompressed volume 800 can be directly used or modified to approximate any similar compressed, decompressed, or uncompressed form / volume of the breast 52 obtained using different imaging modalities for the purpose of registering the corresponding volumes with each other.

[0090] Furthermore, the artifact reduction volume 416 can be used to define the coordinate system of the artifact reduction volume 416. This coordinate system can be employed within any image 420, 422, 424 created from the artifact reduction volume 416, and can be applied by registering the artifact reduction volume 416 with volumes of the breast 52 from other modalities to achieve accurate cross-modal determination of the location of lesions, tumors, or masses between registered volumes.

[0091] Furthermore, the digital uncompressed breast volume 800 generated from the artifact reduction volume 416 can be manipulated to simulate and / or provide a synthetic image and / or volume of the breast 52 obtained using other imaging modalities.

[0092] In yet another exemplary embodiment, the process for obtaining projection 404 in method 400 may be performed as contrast-enhanced DBT or other contrast-enhanced imaging procedures, thereby enhancing the ability of the enhancement processor 32 to distinguish the location of different tissue types or materials within the imaging breast 52 due to the dye (e.g., iodine-based dye) injected into the breast 52 prior to performing the contrast-enhanced imaging procedure.

[0093] Finally, it should be understood that the imaging system 10 and / or any individual computing device used to perform any of the methods in methods 400, 1000, etc., and / or the processes described herein may include necessary electronics, software, memory, storage devices, databases, firmware, logic / state machines, microprocessors, communication links, displays or other visual or audio user interfaces, printing devices, and any other input / output interfaces for performing the functions described herein and / or achieving the results described herein. For example, as previously described, system 10 and / or the individual computing device may include at least one processor and a system memory / data storage structure, which may include random access memory (RAM) and non-transitory read-only memory (ROM). At least one processor of system 10 and / or the individual computing device may include one or more conventional microprocessors and one or more auxiliary coprocessors, such as a math coprocessor. The data storage structure of system 10 and / or the individual computing device discussed herein may include a suitable combination of magnetic, optical, and / or semiconductor memories, and may include, for example, RAM, ROM, flash drives, optical discs such as compact discs, and / or hard disks or drives.

[0094] Additionally, software applications that adapt the controller to perform the methods disclosed herein can read from a computer-readable medium into the main memory of at least one processor. As used herein, the term "computer-readable medium" refers to any medium that provides or participates in providing instructions for execution to at least one processor of System 10 (or any other processor of a separate computing device for performing the methods and / or processes described herein). Such media can take many forms, including but not limited to non-volatile and volatile media. Non-volatile media include, for example, optical, magnetic, or optical disks, such as memory. Volatile media include dynamic random access memory (DRAM), which typically constitutes main memory. Common forms of computer-readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tape, any other magnetic media, CD-ROMs, DVDs, any other optical media, RAM, PROMs, EPROMs or EEPROMs (electronically erasable programmable read-only memory), FLASH-EEPROMs, any other memory chips or cassette tapes, or any other media from which a computer can read.

[0095] Although in the embodiments, execution of a sequence of instructions in a software application causes at least one processor to perform the methods / processes described herein, hardwired circuitry may be used in place of or in combination with software instructions to implement the methods / processes of the present invention. Therefore, embodiments of the present invention are not limited to any particular combination of hardware and / or software.

[0096] It should be understood that the compositions, apparatus, and methods described herein are not limited to the specific embodiments and methods, as these are subject to variation. It should also be understood that the terminology used herein is for the purpose of describing specific exemplary embodiments only and is not intended to limit the scope of this disclosure, which will be limited only by the appended claims.

Claims

1. A method for providing quasi-isotropic images (420, 422, 424) of an object (52), the method comprising the steps of: a. Provide one or more projected images (404) of the object (52) obtained within an angular range of less than 180 degrees relative to the object (52); b. Using artifact reduction and resolution enhancement artificial intelligence (414), a quasi-isotropic volume (416) is reconstructed from the one or more projected images (404). as well as c. Generate one or more quasi-isotropic images (420, 422, 424) from the quasi-isotropic volume (416).

2. The method according to claim 1, wherein the one or more quasi-isotropic images (420, 422, 424) comprises an image plane along at least one plane orthogonal to the detector (18), or comprises an image plane comprising any line between the detector (18) and the focal point of the radiation source (16) of the imaging system (10) for obtaining the one or more projected images (404).

3. The method of claim 2, wherein the one or more quasi-isotropic images (420, 422, 424) include at least one of a plane, a 2D image, a thick slice (420, 432, 434), or a 2D composite image (428), and combinations thereof.

4. The method according to claim 2, wherein the method further comprises the following steps: a. Generate a standard reconstruction volume (410) from the one or more projected images (404); b. Generate one or more standard images (419) from the standard reconstructed volume (410); c. Register the one or more quasi-isotropic images (420, 422, 424) with the standard reconstructed volume (410); as well as d. Synchronize the movement of the cursor (550) in each of the one or more quasi-isotropic images (420, 422, 424) and the one or more standard images (419).

5. The method of claim 3, further comprising the step of employing the one or more quasi-isotropic images in a biopsy procedure.

6. The method of claim 1, wherein the object (52) is a breast (52), and wherein the method further comprises the step of classifying the tissue type of each voxel in the quasi-isotropic volume (416).

7. The method according to claim 6, further comprising the step of numerically decompressing the quasi-isotropic volume (416) into a digital uncompressed volume (800).

8. The method of claim 7, further comprising the step of manipulating the digital uncompressed volume (800) to form an additional view of the breast (52).

9. The method according to claim 7, further comprising the following step: a. Generate a standard reconstruction volume (410) from the one or more projected images (404); b. Register the digital uncompressed volume (800) with the standard reconstructed volume (410).

10. The method according to claim 1, further comprising the following step: a. Generate an uncertainty score (704) for each voxel in the one or more quasi-isotropic images (420, 422, 424); as well as b. Display the uncertainty score (704) together with the one or more quasi-isotropic images (420, 422, 424).

11. A mammography imaging system (10), the mammography imaging system comprising: a. A radiation source (16) that is operable to emit radiation. b. A detector (18) that is alignable with the radiation source (16) and has a surface (50) on which the breast (52) to be imaged is adapted to be positioned; c. A controller (32), operatively connected to the radiation source (16) and the detector (18) to control the operation of the radiation source (16) and the detector (18) to generate image data of the breast (52) in an imaging procedure executed by the imaging system (10), the controller (32) including a central processing unit and an interconnected database, the interconnected database including processor-executable instructions for processing the image data from the detector to create one or more projected images (404) of the breast (52), d. A display (38) operatively connected to the controller (32) to present information to a user; and e. User interface (44), which is operatively connected to the controller (32) to enable user input to the controller (32), The controller (32) is configured to apply resolution-enhanced artificial intelligence (414) to the one or more projected images (404) to reconstruct a quasi-isotropic volume (416) of the breast (52) and generate one or more quasi-isotropic images (420, 422, 424) from the quasi-isotropic volume (416).

12. The imaging system of claim 11, wherein the imaging system (10) is a mammography imaging system (12), the mammography imaging system comprising a compression paddle (28) movable by the controller (32) relative to the detector surface (50) to compress the breast (52) therebetween, and wherein the controller (32) is configured to determine the nature of the tissue forming the lesion (600) in the breast (52) based on a comparison of one or more first quasi-isotropic images (420, 422, 424) obtained under a first compression with one or more second quasi-isotropic images (420, 422, 424) obtained under a second compression, such that the tissue forming the lesion (600) can be distinguished from other breast tissue by determining the mechanical nature of the tissue within the imaged breast (52).

13. The imaging system of claim 11, wherein the controller (32) is further configured to classify the tissue type of each voxel in the quasi-isotropic volume (416).

14. The imaging system of claim 13, wherein the controller (32) is further configured to numerically decompress the quasi-isotropic volume (416) into a digital uncompressed volume (800).

15. The imaging system of claim 13, wherein the controller (32) is further configured to reconstruct a standard reconstruction volume (410) from the one or more projected images (404), generate one or more standard reconstruction images (419) from the standard reconstruction volume (410), register the one or more quasi-isotropic images (420, 422, 424) with the one or more standard reconstruction images (419), and synchronize the movement of a cursor (500) in each of the one or more quasi-isotropic images (420, 422, 424) and the one or more standard reconstruction images (419).

Images