MÉTODO IMPLEMENTADO POR COMPUTADOR, E SISTEMA

Abstract

A computer-implemented method is configured to link each anatomical tissue in a set of anatomical tissues to a corresponding set of possible findings and a different anatomical identifier, map voxels of image data to the different anatomical identifiers, receive a first user input selecting a first anatomical identifier, displaying a first list of possible findings for the first anatomical identifier, receiving a second user input selecting a first finding from the possible findings for the first anatomical identifier, and adding the first finding to a first entry in a list of selected findings.

Description

1 / 27 “COMPUTER-IMPLEMENTED METHOD AND SYSTEM” FIELD OF TECHNIQUE

[001] The following refers generally to a clinical imaging report and, more particularly, to the generation of a structured clinical imaging report and is capable of creating other structured reports. BACKGROUND

[002] The example workflow for a radiology examination includes: a referring clinician prescribing an imaging examination for an individual via an imaging request, a radiology department / center performing the imaging examination on the individual according to the imaging request, the radiology department / center creating a report based on the results at least from a radiologist's interpretation of an image from the imaging examination, and the radiology department / center providing the referring clinician with access to the report. Such reports may be unstructured or structured. An example of an unstructured report is a report generated by a radiologist dictating free text into a recording device and a transcriber transcribing the recording to generate the report. In another example, transcription software (speech-to-text) is used to transcribe the radiologist's free text into the report.An example of a structured report is a digital form with editable report fields that must be populated with entries from predetermined lists of entries for the fields.

[003] With unstructured reports, different clinicians describe what is imaged using different language in different ways to describe the same thing. Structured reports in radiology can lead to a more standardized and therefore quality-controlled report from the radiologist to the referring clinician, which in turn will lead to better-informed treatment decisions and therefore better patient outcomes. However, it can be a challenge for radiologists to keep track of which report fields they have filled in, which report fields have been automatically pre-filled by Petition 870250083979, dated 09 / 18 / 2025, page 49 / 96 2 / 27 a model (e.g., artificial intelligence, etc.), which report field they checked or not, etc. Furthermore, software solutions typically impose a specific workflow, for example, to ensure the integrity of structured reports, rather than supporting well-established diagnostic routines of radiologists. Thus, there is an unmet need for one or more improved approaches to generating a structured report. SUMMARY

[004] The aspects described here cover the problems mentioned above and / or other issues.

[005] In one aspect, a computer-implemented method is configured to link each anatomical tissue in an anatomical tissue set to a corresponding set of possible findings and a different anatomical identifier, map voxels of image data to the different anatomical identifiers, receive a first user command selecting a first anatomical identifier, display a first list of possible findings for the first anatomical identifier, receive a second user command selecting a first finding from the possible findings for the first anatomical identifier, and add the first finding to a first entry in a list of selected findings.

[006] In another aspect, a computer-implemented method is configured to link each anatomical tissue in a set of anatomical tissues to a corresponding set of possible findings and a different anatomical identifier, display an interactive decision tree, in which each section of the decision tree is linked to an anatomical identifier from the anatomical identifiers, receive a first command from the user selecting a section of the decision tree, display a first field of a structured report corresponding to the selected section and a first list of possible findings for a first anatomical tissue corresponding to the first field based on a corresponding first anatomical identifier, and display a subportion of the relevant image data for evaluation. Petition 870250083979, dated 09 / 18 / 2025, page 50 / 96 3 / 27 the first anatomical tissue to select a first discovery from the first list of possible discoveries for the first field.

[007] In another aspect, a system includes a processor that is configured to execute computer executable instructions, which causes the processor to display a user interface to generate a structured report, including displaying structured report fields and displaying image data or the image data and an interactive decision tree, receiving a user command on the image data or the interactive decision tree; and automatically opening a first section of the interactive decision tree corresponding to the selected image data based on the user command on the image data or automatically rendering a region of the image data corresponding to a selected section of the interactive decision tree.

[008] Those skilled in the art will additionally recognize other aspects of the present application after reading and understanding the attached description. BRIEF DESCRIPTION OF THE DRAWINGS

[009] The invention can be implemented in various components and various arrangements of components, and in various stages and arrangements of stages. The drawings are for illustrative purposes only and should not be interpreted as limiting the invention: Figure 1 diagrammatically illustrates an exemplary system, according to one or more embodiments of the present invention; Figure 2 schematically illustrates the example content of a user interface for generating a structured report, according to one or more embodiments of the present invention; Figure 3 illustrates an example of a portion of a decision tree displayed in the user interface, according to one or more embodiments of the present invention; Figure 4 illustrates an example of a portion of a structured report displayed in the user interface, according to one or more embodiments of the present Petition 870250083979, dated 09 / 18 / 2025, page 51 / 96 4 / 27 invention; Figure 5 illustrates an example of a portion of image data displayed on the user interface, according to one or more embodiments of the present invention; Figure 6 illustrates an example of the user interface in connection with a display monitor, according to one or more embodiments of the present invention; Figure 7 illustrates another example of the user interface in connection with a display monitor, according to one or more embodiments of the present invention; Figure 8 illustrates an example of the portion of the structured report with a list of possible options for an editable field of the structured report, according to an embodiment of the present invention; Figure 9 illustrates an example of the image data portion with a slice navigation bar, according to one or more embodiments of the present invention; Figure 10 illustrates an example of the image data portion with an adjacent control slice, according to one or more embodiments of the present invention; Figure 11 illustrates a portion of the example of the decision tree with nodes visually highlighted to indicate a node state, according to one or more embodiments of the present invention; Figure 12 illustrates another part of the example of the decision tree portion with nodes visually highlighted to indicate a node state, according to one or more embodiments of the present invention; Figure 13 illustrates an example of the portion of the structured report in which an unfilled field is filled with an option from the list of possible options, according to one or more embodiments of the present invention; Figure 14 illustrates the part of the example of the decision tree with nodes visually highlighted to indicate a state of the nodes, according to one or Petition 870250083979, dated 09 / 18 / 2025, p. 52 / 96 5 / 27 more embodiments of the present invention; Figure 15 illustrates an exemplary method, according to one or more embodiments of the present invention; Figure 16 illustrates another exemplary method, according to one or more embodiments of the present invention; and Figure 17 illustrates a correlation matrix, according to one or more embodiments of the present invention. DESCRIPTION OF THE MODALITIES

[010] Figure 1 schematically illustrates an exemplary system 102. The system 102 includes a computing system 104, such as a computer, a workstation, etc. The computing system 104 includes a processor 106 and a computer-readable storage medium 108. Non-limiting examples of suitable processors include a central processing unit (CPU), a microprocessor (mP), a graphics processing unit (GPU), and / or another processor. The computer-readable storage medium 108 includes non-transient storage media, such as physical memory, a memory device, etc., and excludes transient media. The computing system 104 additionally includes input / output (I / O) 110.The 104 computing system can be part of a picture archiving and communication system (PACS), an advanced visualization system for radiologists such as Philips® Intellispace® Portal, a cardiovascular information system (CVIS) such as Philips® Intellispace® Cardiovascular or Cardiology PACS (CPACS), a computer workstation, a server and / or other specialized equipment for radiology or cardiology workflow or image reading.

[011] An input device 112 is in electrical communication with the computing system 104 via I / O 110. A non-limiting example of an input device 112 includes a keyboard, a mouse, a microphone, etc. The input device Petition 870250083979, dated 09 / 18 / 2025, page 53 / 96 6 / 27 112 includes one or more input devices. An output device 114 is also in electrical communication with the computing system 104 through I / O 110. A non-limiting example of an output device 114 includes a display monitor, a loudspeaker, etc. The output device 114 includes one or more output devices. In one case, the input device 112 and the output device 114 are separate devices (e.g., a keyboard and a display monitor). In another case, the input device 112 and the output device 114 are the same device (e.g., a touch screen monitor).

[012] In the illustrated embodiment, a remote resource 116 is also in communication with the computing system 104 through I / O 110. The remote resource 116 includes one or more remote resources. Non-limiting examples of the remote resource 116 include an imaging system, a computing and / or archiving system, and / or other resources. Non-limiting examples of the imaging system include a magnetic resonance imaging (MRI) system, a computed tomography (CT) system, an X-ray system, etc. Non-limiting examples of the computing and / or archiving system include cloud processing resources, a server, a workstation, a Radiology Information System (RIS), a Hospital Information System (HIS), an electronic medical record (EMR), a PACS, and / or other computing and / or archiving system.

[013] Processor 106 is configured to execute a computer-readable instruction encoded or embedded in computer-readable storage media 108. At least one computer-readable instruction, when executed by processor 106, causes processor 106 to present a user interface (UI) to generate a structured report. As described in more detail below, in one or more cases, the UI includes an interactive decision tree for the user mapped at least to the fields of the structured report and the image data where a user uses the decision tree, the Petition 870250083979, dated 09 / 18 / 2025, page 54 / 96 7 / 27 structured report and / or image data to trigger the generation of the structured report and, in another case, the interactive decision tree for the user is omitted, closed or not used, the image data is mapped to the fields of the structured report and the user uses the structured report and / or image data to trigger the generation of the structured report.

[014] In one case(s), the use of the decision tree allows visually tracking the progress of the report and allows quick navigation to the most relevant slices for nodes not yet evaluated in the decision tree, which can overcome problems with tracking which report fields have been filled in by radiologists, which report fields have been automatically pre-filled by a template and which report fields have been / have not been checked, etc., and the use of image data allows incorporating the selection of findings needed for the structured report into the well-established (individually) diagnostic workflow, which can overcome solutions that normally impose a certain workflow, for example, to ensure the integrity of structured reports, instead of supporting well-established diagnostic routines of radiologists.

[015] Figure 2 shows a non-limiting example of a UI 202 for generating a structured report. UI 202 includes multiple windows, including window 204, window 206, and window 208. At least one subportion of a decision tree (DT) 210 is displayed in one of the windows 204, 206, and 208, at least one subportion of a structured report (SR) 212 is displayed in another of the windows 204, 206, and 208, and at least one subportion of image data (ID) 214 is displayed in the remaining window of windows 204, 206, and 208. In the illustrated case, window 204 includes decision tree 210, window 206 includes structured report 212, and window 208 includes image data 214. In another case, decision tree 210, structured report 212, and image data 214 are in different windows.

[016] It should be noted that the numbering of the windows (204, 206 and 208) is in order of introduction for explanatory purposes and does not indicate any location. Petition 870250083979, dated 09 / 18 / 2025, page 55 / 96 8 / 27 of any window in UI 202. For example, window 204 does not have to be the leftmost window in the illustrated horizontal arrangement of windows 204, 206, and 208. Furthermore, the illustrated horizontal arrangement of windows 204, 206, and 208 is not limiting. For example, windows 204, 206, and 208 can alternatively be arranged vertically (top to bottom) or some combination thereof (e.g., a horizontal and a vertical arrangement of two). At least two of windows 204, 206, and 208 can also be arranged partially or completely overlapping. In general, the user can place windows 204, 206, and 208 in any desired location relative to each other.

[017] Briefly returning to Figure 3, a subportion 302 of a non-limiting example of decision tree 210 is illustrated. Other style decision trees are also contemplated in the present invention. Subportion 302 includes a set of nodes 304i, ..., 304n (collectively referred to in the present invention as nodes 304), where n is a positive integer greater than or equal to one, corresponding to different anatomy, including Anatomy1, ..., Anatomy.

[018] For each of the nodes 304, the subportion 302 includes a set of branches connected to a set of nodes corresponding to the subanatomy, where a specific anatomy of a node includes subanatomy. For example, the illustrated node 304i includes branches 306i, ..., 306k (collectively called in the present invention branches 306), where k is a positive integer greater than or equal to one, connected respectively to a second set of nodes 308i, ..., 308k (collectively called in the present invention nodes 308), corresponding to Subanatomy1, ..., Subanatomyk.

[019] In some cases, one or more of the subanatomies may be further delineated in different subanatomies (e.g., Subanatomyia, Subanatomyib, etc.). By way of a non-limiting example, the knee (e.g., Anatomyi) includes subanatomies such as menisci (Subanatomyi) and ligaments (Subanatomyj), where the subanatomy of the menisci may be further delineated in the medial meniscus (Subanatomyia) and lateral meniscus (Subanatomyib) and Petition 870250083979, dated 09 / 18 / 2025, page 56 / 96 9 / 27 The subanatomical ligaments can be further delineated into cruciate ligaments (Subanatomyja), medial collateral ligament (Subanatomyjb), and lateral collateral ligament (Subanatomyjc). However, for the sake of brevity, the illustrated subsection 302 shows other sublevels.

[020] For each of the nodes 308, the subportion 302 includes a set of branches connected to a set of nodes corresponding to a binary decision of whether a finding was detected for the subanatomy of the node (e.g., likely normal / no finding and abnormal / no finding). For example, node 308i includes branches 310i and 3102 (collectively referred to in the present invention as branches 310), connected respectively to a set of nodes 312i and 3122 (collectively referred to in the present invention as nodes 312), corresponding respectively to Abnormal and Normal.

[021] For node 312i corresponding to Abnormal, subportion 302 includes a set of branches 314i, ..., 314m (collectively referred to in the present invention as branches 314), where m is a positive integer greater than or equal to one, connected respectively to a set of nodes 316i, ..., 316m (collectively referred to in the present invention as nodes 316), corresponding to a set of different pathologies for Subanatomyi, including Pathologyi, ..., Pathologym.

[022] For each of the nodes 316 corresponding to the set of different pathologies, the subportion 302 includes a set of branches 318i, ..., 318p (collectively called in the present invention branches 318), where p is a positive integer greater than or equal to one, connected respectively to a set of nodes 320i, ..., 320p (collectively called in the present invention nodes 320), corresponding to Graui, ..., Graup.

[023] An example of a knee classification scheme includes the classification scheme for severity grades of osteoarthritis (cartilage) by the International Cartilage Repair Society (ICRS). With this classification, Grade 0 refers to normal cartilage (i.e., intact), Grade 1 refers to nearly normal cartilage with superficial lesions, Grade 2 if Petition 870250083979, dated 09 / 18 / 2025, p. 57 / 96 10 / 27 refers to cartilage with lesions extending to less than 50% of the cartilage depth, Grade 3 refers to cartilage with defects extending to more than 50% of the cartilage depth, and Grade 4 refers to severely abnormal cartilage where the cartilage defects reach the subchondral bone. Other classification schemes are also contemplated in the present invention.

[024] In one case, the displayed subportion 302 of decision tree 210 includes the entire decision tree 210. In another case, the displayed subportion 302 includes less than the entire decision tree 210. For example, the displayed subportion 302 may include the outline as shown in Figure 3, more nodes, or fewer nodes. In one case, the nodes in the displayed subportion 302 of decision tree 210 are based on predetermined display criteria, as described in more detail below.

[025] It should be considered that subportion 302 of decision tree 210 is not limited to a horizontal flow (nodes 304,..., 320), from left to right, with subnodes within any specific node being arranged vertically (nodes 3041,..., 304n). In another case, subportion 302 of decision tree 210 is arranged vertically from top to bottom flow. In this case, the subnodes 304 of at least one node may be arranged horizontally, vertically and / or otherwise. In another case, nodes 304,..., 320 may be arranged with a combination of horizontal and vertical flow.

[026] Briefly returning to Figure 4, a non-limiting example of a subportion 402 of the structured report 212 is illustrated. Similar to the displayed subportion 302 of the decision tree 210, subportion 402 may include the entire structured report or less than the entire structured report. For explanatory purposes and for brevity, the illustrated subportion 402 includes fields corresponding to a trajectory of the subportion 402 of the structured report 212 shown in Figure 3, namely, from 304i to 308i and 312i to 316 and 320.

[027] For example, the illustrated subsection 402 includes editable report fields. 404, 406, 408, 410 and 412, which correspond respectively to nodes 304, 308, 312, 316 and 320 in Figure 3, which are identified respectively at 414, 416, 418, 420 Petition 870250083979, dated 09 / 18 / 2025, p. 58 / 96 11 / 27 and 422. In the illustrated subsection 402, entries were selected for editable report fields 404, 406, and 408, but not yet for editable report fields 410 and 412. As discussed in greater detail below, entries can be selected based on user command selecting an entry from a list of possible entries and / or automatically based on the results of a machine learning or other algorithm.

[028] Briefly returning to Figure 5, a non-limiting example of a subportion 502 of the image data 214 is illustrated. Similar to the displayed subportion 302 of the decision tree 210 and the displayed subportion 402 of the structured report 212, subportion 502 may include all image data (e.g., a three-dimensional (3D) rendering and / or one or more visible two-dimensional (2D) images / slices thereof) or less than all image data (e.g., a 3D rendering and / or one or more visible 2D images / slices corresponding to the decision tree subportion 210 and / or structured report 212). The illustrated subportion 502 includes a slice relevant to the editable report field 410 in Figure 4.

[029] In one case, windows 204, 206, and 208 are displayed on a single display monitor. In another case, windows 204, 206, and 208 are displayed on a different display monitor. In yet another case, two of windows 204, 206, and 208 are displayed on one display monitor, and the remaining window of windows 204, 206, and 208 is displayed on a different display monitor. One or more of windows 204, 206, and 208 can be configured to be minimized and / or closed or never opened. For example, in one case, a user can, through an input (e.g., mouse, keyboard, voice activation, gesture, etc.), minimize a window where the window content is no longer visible and a graphic such as an icon, etc., representing the window is displayed on the screen in place of the window.

[030] By way of non-limiting example, Figure 6 shows a single monitor 602 with windows 206 and 208 open, but not maximized as they do not cover the entire display area 604 and a graph 606 representing the window Petition 870250083979, dated 09 / 18 / 2025, page 59 / 96 12 / 27 minimized 204. In Figure 7, window 204 is either closed or never opened. For example, if a user decides not to use decision tree 210, the user can close the window displaying decision tree 210 or never open the window displaying decision tree 210 in the first place. In one case, the user can open a window that is either closed or never opened. Alternatively, Figure 7 shows a case where decision tree 210 is not part of the structured report generation software and is not displayed and cannot be used.

[031] With further reference to Figures 1 to 7, as briefly discussed above, the displayed sub-portions 302, 402 and / or 502 may include less than the entire decision tree 210, the entire structured report 212 and / or all image data 214. In one case, the displayed information is based on a standard, a user preference, a healthcare entity preference and / or otherwise. For example, the complete decision tree 210 may be quite detailed and a user may prefer only a partial view. In that case, downstream nodes may not be shown until they are needed, for example, in connection with Figure 3, classifications may not be displayed until a pathology is selected and then only the classifications for the selected pathology are displayed.

[032] In another case, interaction with one of the displayed subportions 302, 402 and / or 502 of the decision tree 210, the structured report 212 and / or the image data 214 controls what is displayed in the other subportions 302, 402 and / or 502 of the decision tree 210, the structured report 212 and / or the image data 214. For this, an anatomical model is used to link the decision tree 210, the structured report 212 and / or the image data 214 together. In one case, the anatomical model(s) provide(s) a multi-label output, for example, to be able to segment a multitude of anatomical structures. In another case, the anatomical model(s) provide(s) a combination of multiple segmentation models that provide single-label outputs or a combination thereof. In another case, the anatomical model(s) provide(s) both or a combination thereof. Petition 870250083979, dated 09 / 18 / 2025, pp. 60 / 96 13 / 27 of the same.

[033] A suitable anatomical model provides a map from an image voxel to a set of anatomical identifiers (IDs), for example, voxelxyz includes tissue anatomyi or anatomysi-n, where the anatomical identifiers map to the same anatomyi or anatomysi-n in the decision tree 210 and / or structured report 212. Examples of suitable models include U-NET or F-NET based convolutional neural networks (CNNs) and the like. Other suitable examples are discussed in EP3493154, filed on 12 / 01 / 2017 and entitled "Segmentation system for segmenting an object in an image", US9824457B2, filed on 08 / 21 / 2015 and entitled "Model-based segmentation of an anatomical structure", EP 3120323, filed on 02 / 27 / 2015 and entitled "Image processing apparatus and method for segmenting a region of interest", all of which are incorporated herein in their entirety by reference.

[034] By way of non-limiting example, in one case, user interaction with structured report 212 determines displayed subportion 302 of decision tree 210. For example, in one case, when a user selects (e.g., clicks, hovers over, points to, verbally identifies, etc.) a section of structured report 212, displayed subportion 302 of decision tree 210 corresponds to the anatomical tissue covered in the section of structured report 212. Additionally or alternatively, displayed subportion 502 of image data 214 corresponds to the anatomical tissue covered in the section of structured report 212.

[035] In another case, the user's interaction with image data 214 determines the displayed subportion 302 of the decision tree 210. For example, in one case, when a user selects (e.g., clicks, hovers over, points to, verbally identifies, etc.) a section (e.g., an image, a set of images, a region of a 3D rendering, etc.) of image data 214, the displayed subportion 302 of the decision tree 210 corresponds to the anatomical tissue covered in the section of image data 214. Additionally or alternatively, the displayed subportion 502 of the structured report 212 corresponds Petition 870250083979, dated 09 / 18 / 2025, pp. 61 / 96 14 / 27 to the anatomical tissue in image data section 214.

[036] In another case, the user interaction with the displayed subportion 302 of the decision tree 210 determines the displayed subportion 302 of the structured report 212 and / or the displayed subportion 502 of the image data 214. For example, in one case, the displayed subportion 502 of the image data 214 includes a slice(s) that intersect a mesh center of mass, voxels classified as structure X in response to a user clicking on a decision tree node of the displayed subportion 302 of the decision tree 210 and / or the displayed subportion 302 of the structured report 212 related to structure X, etc.

[037] The following describes examples of user interaction with UI 202 using image data 214 to trigger structured report generation, with or without decision tree availability 210.

[038] In one case, a user command selects a portion of the displayed image data 214. In response to this, the processor 106 displays a section of the structured report 212 that covers the anatomical tissue in the selected portion of the displayed image data 214 along with a list of possible findings for the anatomical tissue. As discussed in this document, a set of possible entries, including findings, can be presented in a drop-down list, a pop-up menu, a list box, and / or otherwise. Figure 8 shows an example of a list 800 presented in response to the selection of a portion of the displayed image data 214. The list includes options 802, 804, and 806. This same list 800 and / or another list can be invoked and presented in response to user interaction with the decision tree 210.

[039] In one case, the displayed list contains only findings that are linked to the anatomical IDs contained in the user-selected image portion (for example, a cruciate ligament tear is only available for selection when the cruciate ligament is contained in the image portion). In another case, findings linked to anatomical IDs within a predetermined neighborhood (for example, within x centimeters (cm)) of the image portion are suggested. Petition 870250083979, dated 09 / 18 / 2025, pp. 62 / 96 15 / 27 user-selected image. In one instance, this could explain imprecise user segmentation / interaction errors, for example, tibial and femoral cartilage defects might be displayed when the user selects a voxel that, according to the anatomical model, belongs to the neighboring meniscus.

[040] In one instance, the user-selected portion of the image data is selected by clicking on voxels or slices, freehand drawing of a bounding box, placement and sizing of a predetermined bounding box, etc. Additionally or alternatively, the user-selected portion of the image data is selected by hovering a pointing device, such as a mouse, etc., over a voxel, an area, and / or a slice, for example, for a predetermined period of time. In one instance, voxels that are connected to anatomical tissue already covered in the report are marked as such (e.g., with a color, an icon, etc.) so that the clinician knows that the anatomical tissue is already included in the structured report 212.

[041] In one case, in a separate viewing window, a list of all anatomical tissues is displayed, and the appearance of the anatomical tissues depends on the classification (e.g., not visualized = source 1 characteristic, normal = source 2 characteristic, abnormal = source 3 characteristic). In one case, the appearance further depends on whether the finding is a suggestion by a classification model, confirmed, or altered by the clinician, e.g., source A characteristic: model suggestion, source B characteristic: selected by the radiologist).

[042] In another case, a 3D rendering of the anatomical tissues is displayed and, similarly, the rendering of the anatomical tissues (e.g., transparency, color) depends on the classification situation. In one case, a clinician can click on an anatomical tissue in the list / rendering and the displayed image data 214 automatically includes the most relevant sequence(s) and / or slice(s) for the corresponding anatomical tissue (e.g., based on the anatomical model, e.g., a slice containing the center of mass or the largest diameter etc. of the tissue). Petition 870250083979, dated 09 / 18 / 2025, pp. 63 / 96 16 / 27 anatomical, or based on a prominence map for the suggested finding) is displayed. In one case, an anatomical tissue is classified as not seen if the clinician has not seen all the relevant slices to evaluate that structure, which can be predefined, for example, it can be defined that the meniscus must at least be seen in coronal view in at least one proton density (PD) sequence.

[043] In one case, the list of possible discoveries displayed by the processor 106 corresponding to the selected portion of the image data 214 includes all findings that are linked to the anatomical IDs contained in the user-selected image portion without further filtering / sorting. In another case, the entries in the list are sorted according to a classification model. For this, the model can be a statistical model. In one case, the model output is a statistical prevalence (e.g., based on historical data) of the findings.

[044] In one instance, in response to a clinician not selecting any voxel from the displayed image data 214, the anatomical tissue in the displayed image data 214 is automatically classified as normal / no finding in the structured report 212. In another instance, when the clinician has already decided on a finding in a specific anatomical region, the subject's finding and / or findings are excluded because the selected finding is removed from the list displayed for additional user interaction. For example, if the clinician's command selects an entry that identifies a ligament as torn, an entry for a strained ligament is no longer a possible entry and is automatically removed from the displayed list.

[045] In one case, the decision to automatically remove a possible entry is based on rules or a rule. For example, continuing with the example above, a rule could indicate that when torn is selected as the entry for the ligament, non-torn ligament states such as strained, bruised, etc. are automatically removed from the displayed list. The rule could be Petition 870250083979, dated 09 / 18 / 2025, pp. 64 / 96 17 / 27 is invoked in response to the clinician's command selecting the entry that identifies the ligament as torn and / or otherwise. In one case, the clinician can undo the selection. In that case, the automatically removed entry is automatically added back to the list of possible entries and can be selected by the clinician.

[046] In a case where a classification model is used to rank the list of possible findings, several user interactions are contemplated. For example, in one case, a more likely finding is automatically selected when the clinician no longer interacts with the list of possible findings and moves (e.g., scrolls, etc.) to the next slice. In another case, the clinician is warned when a manual selection does not match the output of the classification model, for example, the classification model identified an abnormality (e.g., a certain probability threshold for an abnormality is exceeded) in an anatomical structure, but the radiologist did not click on any voxel of the anatomical structure or vice versa. This warning may be a pop-up warning or the classification model's proposed finding may be displayed in a different color.

[047] In one case, the window (and / or a separate window) in which the displayed subportion 502 of the image data 214 is displayed includes a slice navigation bar with graphical cues (e.g., color highlighting, icons, etc.) that identify a slice(s) of the image data 214 displaying anatomical tissues for which the classification model has identified abnormal findings, so that the radiologist can quickly navigate to slices with possible abnormal findings. In another case, in addition to a sorted list, an indication of which area / region of interest (ROI) led to the abnormal classification is shown to the user, for example, by a colored overlay on the image or a bounding box.

[048] Figure 9 shows an example with a slice navigation bar 900 to one side of the displayed subportion 502 of the image data 214. A current location marker 902 indicates a location of the displayed slice 502 within the data. Petition 870250083979, dated 09 / 18 / 2025, pp. 65 / 96 18 / 27 image. A relevant slice location marker 904 indicates a slice in image data 214 with anatomical tissues for which the classification model identified abnormal findings. When the user clicks marker 904, portion 502 of image data 214 displays the slice. Figure 10 shows an alternative case where a Next control 1002 is displayed in the UI and the user invokes the Next control to display the slice in image data 214. Another case includes slice navigation bar 900 and Next control 1002.

[049] The following describes examples of user interaction with UI 202 using decision tree 210 to trigger structured report generation.

[050] In one instance, the displayed subportion 302 of decision tree 210 is visually interactive. For example, in one instance, a feature of a source from one or more nodes of decision tree 210 visually conveys information about one or more nodes. Examples of suitable features include, but are not limited to, a color (e.g., a hue, a tone, a tint, etc.), a style (e.g., regular, italic, bold, bold italic, etc.), an effect (e.g., all uppercase, lowercase, etc.), a size, a background, a transparency / opacity, etc.

[051] As a non-limiting example, the source characteristic of one or more nodes in decision tree 210 may identify a discovery availability. For example, in one case, a source characteristic for an available discovery may include one characteristic and a source characteristic for an unavailable discovery may include a different characteristic. For example, a source color for an available discovery may be color 1 and a source color for an unavailable discovery may be color 2, where color 1 and color 2 are different colors.

[052] For example, Figure 11, which shows a subportion of Figure 3, indicates that the Discovery field 418 in the displayed portion 402 of the structured report 212 shown in Figure 4 is filled, but the Pathology field 410 is empty, using a lighter shade of gray for the text in 3121 in the decision tree 210 and a shade Petition 870250083979, dated 09 / 18 / 2025, pp. 66 / 96 19 / 27 of darker gray for the text in fields 316i.....316m in decision tree 210. In Figure 4, editable field 408 for the Discovery field 418 is filled with Abnormal and editable field 410 for the Pathology field 420 is unfilled / empty.

[053] In one instance, the font characteristic changes in response to a clinician entering a finding in an unfilled field. Figure 12 shows that the text in field 3161 changes to the same lighter shade of gray as the text in 3121 in decision tree 210 in response to the user selecting a pathology for editable field 410 for Pathology field 420, and Figure 13 shows the structured report from Figure 4 updated to reflect that a pathology (i.e., Pathologyt) was selected for editable field 410 for Pathology field 420.

[054] In one case, a source characteristic changes based on a discovery state. For example, in one case, a source characteristic changes in response to a clinician confirming or rejecting an auto-populated discovery. In one case, the auto-populated discovery is a discovery that was determined based on artificial intelligence (AI), such as machine learning (ML). For example, in one case, the discovery is determined using a neural network, such as a convolutional neural network, that was trained on an image set with discoveries where the output is a set of probabilities for each discovery. When the probability of an AI-determined discovery meets a predetermined threshold, the AI-determined discovery is auto-populated, and when the probability of an AI-determined discovery does not meet the predetermined threshold, the AI-determined discovery is not auto-populated.

[055] In a case where editable fields 408 and 410 in Figure 13 are automatically filled in as such, Figure 14 shows the text in 3121 in decision tree 210 in italics, indicating that the user accepted the automatically filled input, and the text in 3161 in decision tree 210 underlined, indicating that the user rejected the automatically filled input. In general, the change Petition 870250083979, dated 09 / 18 / 2025, pp. 67 / 96 20 / 27 of the font characteristic is rule-based or based on a single rule. For example, in the example above, a rule would state that if an AI-determined, auto-populated finding is accepted by a clinician, the entry text will be displayed in italics. The same rule, or another rule, would state that if the AI-determined, auto-populated finding is rejected by the clinician, the entry text will be underlined.

[056] Additionally or alternatively, a source feature depends on an automatically populated input certainty, for example, based on a cross-entropy of classification results, for example, where one feature indicates high certainty while a different feature indicates low certainty. Additionally, or alternatively, different source features indicate different information, for example, a situation of a decision tree node assessment (e.g., bold = verified by radiologist and normal = not yet verified) while a color indicates availability and / or certainty of discovery.

[057] In one instance, in response to a user selecting a node from the decision tree 210, processor 106 displays a set of possible entries for the node. For example, where the selected node is pathology node 3161 in Figure 14, which was populated with Pathology1 as shown in Figure 13, a set of possible entries for the node is displayed with the structured report 212 for the editable field 412 for the Grade field 422 (similar to Figure 8). The window may include a dropdown menu, a pop-up menu, a list box, and / or other form. A user selects an entry from the list to populate the field.

[058] When possible entries are discovered, the entries can be classified, for example, based on statistical prevalence determined from historical data, according to an output of a classification model. In another case, a more likely finding is automatically selected for a node when the clinician does not interact with the possible entries associated with the node and instead selects another node from the decision tree 210. Alternatively, the Petition 870250083979, dated 09 / 18 / 2025, pp. 68 / 96 The 21 / 27 discovery is automatically set to normal, and a warning is displayed to inform the user when the most likely discovery was not normal.

[059] In one instance, an unpopulated node of the 210 decision tree upstream of another unpopulated node of the 210 decision tree is automatically populated in response to the downstream node of the 210 decision tree being populated. For example, if a clinician selects a grade 3 meniscal tear of the anterior horn of the lateral meniscus in the downstream part of the 210 decision tree, then the upstream decisions abnormal / normal meniscus?, Which part of the meniscus? and Above grade 1? are implicitly made by the clinician and are automatically populated and set as checked and selected by the clinician.

[060] In one case, after the fields of a node have been completed, the user selects a next node in the decision tree 210 to complete. Alternatively, processor 106 recommends a next node for the user to complete. In one case, processor 106 is triggered to provide a recommendation based on a user command, for example, a user activating a Finish control button on the screen with a mouse, a pen (e.g., a stylus, a finger, etc.), a gesture, a verbal command, etc., or similarly, a Next control button. In one case, the recommended node is highlighted, for example, visually through a flashing light, a color change, etc., audibly through a speaker under running speech software, etc. In another case, the recommendation causes the node to be automatically selected and opened for interactive evaluation.

[061] In one instance, processor 106 recommends a next node to report on based on a maximum correlation with the findings in the current node. As a non-limiting example, after the anterior horn of the lateral meniscus has been completed, processor 106 recommends the pars intermedia of the lateral meniscus, since the findings in these two areas are highly correlated. In one instance, such correlations can be calculated by analyzing a collection of sample findings, for example, from a population of individuals with an equal finding. Petition 870250083979, dated 09 / 18 / 2025, pp. 69 / 96 22 / 27 or similar. In another example, processor 106 recommends a next node to report on based on a history of the clinician's notes, for example, recommending a node that the clinician has chosen most frequently.

[062] By way of a non-limiting example, with respect to maximum correlation, in one case, the available clinical records from a database, a server, a RIS, a HIS, an EMR, a PACS, etc. are analyzed to identify a population of individuals with an identical or similar finding. With respect to the example above, the identical or similar finding would be a finding for the anterior horn of the lateral meniscus. The identified population is analyzed to identify other related findings. In the example above, the other related findings would include findings related at least to the meniscus, such as a part of the meniscus adjacent to the anterior horn, for example, the intermediate part of the lateral meniscus.

[063] A statistical analysis is performed to determine a probability, a possibility, etc. with which each of the other related findings occurs in the clinical records of the identified population. The maximum correlation corresponds to the findings with the highest probability, possibility, etc. Processor 106 recommends the node corresponding to the maximum correlation. With respect to the example above, the maximum correlation corresponds to the findings for the pars intermedia of the lateral meniscus, and the node corresponding to it is recommended as the next node. Other examples of correlated findings include bone marrow edema and cartilage damage, anterior cruciate ligament rupture and a bone hematoma, etc.

[064] Such correlations between results can be predetermined and stored in a look-up table (LUT), a matrix, a bar chart, and / or in another form. Briefly returning to Figure 17, an example of a 1702 matrix of correlations between findings for a set of findings is illustrated. It should be understood that the set of findings in the 1702 matrix is ​​not limiting; other correlation matrices may include more or fewer findings, Petition 870250083979, dated 09 / 18 / 2025, pp. 70 / 96 23 / 27 additional and / or different findings, etc. The 1702 correlation matrix shows the set of findings (on a first axis 1704) versus the same set of findings (on a second axis 1706).

[065] Each intersection of two findings in the 1702 correlation matrix indicates a correlation between the two findings, and a correlation value between two findings is represented as a gray level of a 1708 gray scale that is mapped to correlation values. For example, a 1710 finding of meniscus_tear intersects with a 1712 finding of bone_trauma at 1714 and intersects with a 1716 finding of cartilage_any_defect at 1718. Each of the 1714 and 1718 intersections is represented by a gray level of the 1708 gray scale. The 1710 finding of meniscus_tear also intersects with other findings, which are not discussed in detail.

[066] In this example, from grayscale 1708, discovery 1710 of meniscus_tear is more positively correlated with discovery 1716 of cartilage_any_defect than with discovery 1712 of bone_trauma, as the grayscale level in 1718 for discovery 1716 of cartilage_any_defect maps to a higher grayscale level (around 0.27) in grayscale 1708, and the grayscale level in 1714 for discovery 1712 of bone_trauma maps to a lower grayscale level (around 0.00) in grayscale 1708. Discovery 1710 of meniscus_tear crosses with other discoveries in matrix 1702 with correlation values ​​between the correlation values ​​for bone_trauma and cartilage_any_defect.

[067] With matrix 1702, after the clinician enters meniscus_tear finding 1710 into the input for the current node, processor 106 determines that cartilage_any_defect finding 1716 maps to a higher positive correlation for findings in matrix 1702 and recommends, as the next node, the node corresponding to cartilage_any_defect finding 1716.

[068] In another case, an AI-based method is used to identify a typical disease pattern. Disease patterns can be defined as a Petition 870250083979, dated 09 / 18 / 2025, p. 71 / 96 24 / 27 collection of findings that accompany, for example, a typical type of injury (e.g., knee sprain, twisted knee). The next node for reporting can be derived from the typical collection of findings. The AI-based method can be trained by having a database of a specific type of imaging exams (e.g., knee MRI) and assigning typical patterns to the exams in the database. When the AI ​​algorithm is applied to an exam of that specific type, the AI ​​algorithm can classify the exam against the patterns, for example, twisted knee exams usually include a finding of a knee sprain. In this way, the 106 processor recommends the node that corresponds to knee sprains as the next node.

[069] In another case, collaborative filtering can be used to identify individuals in a dataset who have a similar pathological structure. As a non-limiting example, when a clinician confirms a meniscus tear, the 106 processor can search for other individuals in the dataset who have a meniscus tear and identify other pathologies that these individuals have in common with the individual under evaluation. The probability for these findings can be determined, and the node corresponding to the finding with the highest occurrence can be recommended as the next node.

[070] It should be considered that one or more (including all) of the previously mentioned features may be automated, for example, performed by the system without a user command. However, to the extent that a user desires more control, one or more (including all) of the previously mentioned features may be semi-automated, requiring a user command, for example, a user command accepting / rejecting an automated action.

[071] Figure 15 reveals a computer-implemented method. It should be noted that the order of actions in one or more of the methods is not limiting. Thus, other sequences are contemplated in the present invention. Furthermore, one or more actions may be omitted and / or one or more additional actions may be included. A linking step 1502 links each anatomical tissue into a set of Petition 870250083979, dated 09 / 18 / 2025, pp. 72 / 96 25 / 27 anatomical tissues to a corresponding set of possible findings and a different anatomical identifier. A mapping step 1504 maps voxels of image data to the different anatomical identifiers. A receiving step 1506 receives a first command from the user selecting a region of the image data. A display step 1508 displays a first field of a structured report with a first list of possible findings for the first anatomical tissue in the selected region based on a corresponding first anatomical identifier. A receiving step 1510 receives a second command from the user selecting a first finding from the first list of possible findings for the first anatomical tissue. A filling step 1512 fills the first field of a structured report corresponding to the first anatomical tissue with the first selected finding.

[072] Figure 16 reveals another computer-implemented method. It should be considered that the order of actions in one or more of the methods is not limiting. Thus, other sequences are contemplated in the present invention. In addition, one or more actions may be omitted and / or one or more additional actions may be included. A linking step 1602 links each anatomical tissue in a set of anatomical tissues to a corresponding set of possible findings and a different anatomical identifier. A display step 1604 displays an interactive decision tree to the user, where each section of the decision tree is linked to an anatomical identifier from the anatomical identifiers. A receiving step 1606 receives a first command from the user selecting a section of the decision tree.A display step 1608 displays a first field of a structured report corresponding to the selected section and a first list of possible findings for a first anatomical tissue corresponding to the first field based on a first corresponding anatomical identifier. A display step 1610 displays a subportion of the image data relevant to evaluating the first anatomical tissue to select a first finding from the first list of possible findings for the first field. Petition 870250083979, dated 09 / 18 / 2025, pp. 73 / 96 26 / 27

[073] The above methods can be implemented by means of computer-readable instructions, encoded or embedded in computer-readable storage media, which, when executed by a computer processor, cause the processor to perform the actions or functions described. Additionally or alternatively, at least one of the computer-readable instructions is carried by a signal, a carrier wave or other transient media, which is not computer-readable storage media.

[074] Although some examples in this document have been provided in the field of radiology, the systems, methods and / or operations described in this document can be implemented in other areas where structured reporting is used, such as cardiology, etc. Furthermore, although some examples in this document are described in connection with an imaging examination of the knee, the systems, methods and / or operations described in this document can be implemented for other anatomy and / or organs, such as the heart, for example, in the case of a CVIS or cardiology workflow.

[075] Although the invention has been illustrated and described in detail in the aforementioned drawings and description, such illustration and description should be considered illustrative or exemplary, and not restrictive; the invention is not limited to the embodiments disclosed. Other variations of the embodiments disclosed may be understood and performed by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the appended claims.

[076] The expression that includes does not exclude other elements or other steps, and the indefinite article a or an does not exclude a plurality. A single processor or other unit may perform the functions of several items mentioned in the claims. The mere fact that certain measures are mentioned in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

[077] A computer program can be stored / distributed in a Petition 870250083979, dated 09 / 18 / 2025, pp. 74 / 96 27 / 27 suitable media, such as optical storage media or solid-state media, supplied together with or as part of other hardware, but may also be distributed in other ways, such as via the Internet or other wired or wireless telecommunication systems. Any references in the claims should not be construed as limiting the scope. Petition 870250083979, dated 09 / 18 / 2025, pp. 75 / 96

Claims

1 / 6 CLAIMS 1. A computer-implemented method characterized by comprising: linking each anatomical tissue in a set of anatomical tissues to a corresponding set of possible findings and a different anatomical identifier; displaying a decision tree to a user, wherein each section of the decision tree is linked to an anatomical identifier from the anatomical identifiers; receiving a first command from the user selecting a section of the decision tree; displaying a first field corresponding to the selected section and a first list of possible findings for a first anatomical tissue that corresponds to the first field based on the first corresponding anatomical identifier; and displaying a subportion of the image data relevant to evaluating the first anatomical tissue to select a first finding from the first list of possible findings for the first field.

2. Method according to claim 1, characterized by further comprising: mapping voxels of image data to different anatomical identifiers; receiving a second command from the user selecting a region of the image data; displaying a second field with a second list of possible findings for the second anatomical tissue in the selected region based on a corresponding second anatomical identifier; receiving a third command from the user selecting a second finding from the second list of possible findings for the second anatomical tissue; and adding the second finding to a first entry in a list of selected findings.

3. A method, according to any one of claims 1 or 2, characterized by further comprising: the identification section of the decision tree including a field with an automatically populated discovery with a first visual highlight; and changing the first visual highlight to a second visual highlight in response to a user accepting or rejecting the automatically populated discovery.

4. Method, according to any one of claims 1 to 3, characterized by further comprising: identifying a certainty of discovery automatically filled with a third visual highlight.

5. A method, according to any one of claims 1 to 4, characterized by further comprising: identifying the decision tree section and including an unfilled field with a fourth visual highlight; and changing the fourth visual highlight to a fifth visual highlight in response to a user adding a discovery to the unfilled field.

6. Method, according to any one of claims 1 to 4, characterized by further comprising: automatically displaying a next field based on a predetermined correlation with the findings in the selected section.

7. A method, according to any one of claims 1 to 4, characterized by further comprising: automatically displaying a next field based on a user's annotation history.

8. Method, according to any of the preceding claims, characterized in that a structured report (212) is generated based on the list of possible findings for the first field, wherein the user has the possibility of selecting a section of the structured report and wherein a displayed subportion of the decision tree corresponds to the anatomical tissue represented in the section of the structured report.

9. Method according to claim 8, characterized in that the discovery field (418) is displayed in the structured report configured to include automatically filled entries, wherein the system user can accept or reject the entries.

10. Method, according to claim 9, characterized in that the discovery field includes nodes and the user has the possibility to select nodes and that through the selection, a set of possible entries for the node are displayed based on a rule-based approach, depending on the acceptance or rejection of the entries by the user.

11. A method according to claim 10, characterized in that an unpopulated node of the decision tree 210 upstream of another unpopulated node of the decision tree 210' is automatically populated in response to the downstream node of the decision tree 210 being populated.

12. A method, according to any of the preceding claims, characterized by, instead of using a decision tree, mapping voxels of image data to different anatomical identifiers, with an additional step of adding the first finding to the first entry in a list of selected findings being performed.

13. A method, according to claim 12, characterized by hovering a pointing device over a voxel, an area and / or a slice, for example, for a predetermined period of time, is highlighted, and the voxels that are connected to the anatomical tissue already addressed in the report are marked as such so that the physician knows that the anatomical tissue is already included in the structured report.

14. Method, according to any of the preceding claims, characterized by the 3D rendering of anatomical tissues being displayed based on Petition 870250083979, dated 09 / 18 / 2025, page 93 / 96 4 / 6, depending on the classification state of the anatomical tissue, such that, depending on the user's command, the displayed image data automatically includes the sequence(s) and / or slice(s) most relevant to the corresponding anatomical tissue.

15. Method, according to any one of claims 12 to 14, characterized in that, in response to the physician not selecting any voxel from the displayed image data 214, the anatomical tissue from the displayed image data 214 is automatically classified as “normal / no findings” in the structured report 212.

16. A method, according to any of the preceding claims, characterized by a classification model being used to classify the list of possible findings, with the most probable finding being automatically selected when the physician no longer interacts with the list of possible findings and / or slices for a certain period of time.

17. Method, according to any one of claims 12 to 16, characterized in that an anatomical model is used to link the decision tree 210, the structured report 212 and / or the image data 214, providing a multiple label output.

18. System (102) characterized by comprising: a memory (108) with computer executable instructions; and a processor (106) configured to execute the computer executable instructions that cause the processor to: display a user interface to generate a structured report, including displaying structured report fields and displaying image data or the image data and an interactive decision tree to the user, receive a user command on the image data or on the user interactive decision tree; and automatically open a first section of the user interactive decision tree corresponding to the selected image data based on the user command on the image data or automatically render a region of the image data corresponding to a second selected section of the interactive decision tree to the user.

19. System, according to claim 18, characterized in that the computer executable instructions additionally cause the processor to: link each anatomical tissue in a set of anatomical tissues to a corresponding set of possible findings and a different anatomical identifier; display an interactive decision tree to the user, wherein each section of the decision tree is linked to an anatomical identifier from the anatomical identifiers; receive a first command from the user selecting a section of the decision tree; display a first field of a structured report corresponding to the selected section and a first list of possible findings for a first anatomical tissue corresponding to the first field based on a corresponding first anatomical identifier;and display a sub-portion of the relevant image data to evaluate the first anatomical tissue in order to select a first finding from the first list of possible findings for the first field.

20. System, according to claim 18, characterized in that the computer executable instructions additionally cause the processor to: link each anatomical tissue in a set of anatomical tissues to a corresponding set of possible findings and a different anatomical identifier; map voxels of image data to the different anatomical identifiers; receive a first command from the user selecting a region of the image data; Petition 870250083979, dated 09 / 18 / 2025, p.95 / 96 6 / 6 display a first field of a structured report with a first list of possible findings for the first anatomical tissue in the selected region based on a first corresponding anatomical identifier; receive a second command from the user selecting a first finding from the first list of possible findings for the first anatomical tissue; and populate the first field of a structured report corresponding to the first anatomical tissue with the first selected finding. Petition 870250083979, dated 09 / 18 / 2025, pp. 96 / 96.

Images