User assist for vascular analysis and modeling

Abstract

A method for vascular modeling including accessing a plurality of medical image data depicting a portion of a vasculature of a subject comprising one or more vessels. A method for vascular modeling may display, via a user interface, at least one medical image of the plurality of medical image data. A method for vascular modeling may annotate, via the user interface, at least one indicator of the one or more vessels of the subject of the displayed at least one medical image. A method for vascular modeling may concurrently display, via the user interface, at least one prompt on the at least one medical image by the system, wherein each of the at least one prompts corresponds to user instructions for annotating a specific indicator of the at least one indicators.

Description

USER ASSIST FOR VASCULAR ANALYSIS AND MODELINGCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Prov. Patent App. No. 63 / 666035 titled “USER ASSIST FOR VASCULAR ANALYSIS AND MODELING” and filed on June 28, 2024, and U.S. Prov. Patent App. No. 63 / 799260 titled “USER ASSIST FOR VASCULAR ANALYSIS AND MODELING” and filed on May 2, 2025, the disclosures of which are hereby incorporated herein by reference in their entirety.BACKGROUND

[0002] Cardiovascular disease (CVD) is a leading cause of morbidity and mortality, with an estimated 244.1 million people worldwide with CVD, particularly due to the subsect of CVD, coronary artery disease (CAD). CAD can include acute coronary syndromes (ACS) and stable angina pectoris (SAP). CAD may involve a prolonged asymptomatic developmental phase, with clinical manifestations that often result in angina pectoris, acute myocardial infarction (MI), or cardiac death. The underlying mechanism that may cause CAD involves atherosclerotic lesions of the coronary arteries. Atherosclerosis is a plaque buildup that narrows the coronary arteries and decreases blood flow to the heart, resulting in ischemia or coronary stenosis. Revascularization is the preferred therapy for patients with moderate to severe ischemia or stenosis, resulting in significant improvements for the patient due. Revascularization strategies include many techniques such as open-heart surgery, coronary artery bypass grafting (CABG), and percutaneous coronary intervention (PCI) methods such as balloon angioplasty, bare-metal stents (BMS), and first- and second-generation drug-eluting stents (DES). The severity of CAD can be assessed through vascular computer models.

[0003] Coronary physiology assessments, also known as lesion functional assessments, provide an indication of the health of the cardiac vasculature of a patient by providing characteristics of the vessels. These characteristics can be lesions that may be an indication of whether atherosclerosis or other forms of CAD are present. Vascular assessments, such as quantitative coronary angiography (QCA), can be a minimally invasive coronary physiology assessment that can determine the characteristics of the cardiac vasculature using medical images obtained during a procedure. A number of these cardiac images (e.g., angiographic images, also referred to herein as angiograms) may be obtained that depict a portion of apatient’s heart. These cardiac images may represent angiographic images and may be obtained through minimally invasive procedures, such as a catheter inserted into the artery, or non- invasive procedures or assessments, such as an x-ray or computer tomography (CT) scan. These cardiac images can be 2-dimensional (2D) angiographic images.

[0004] These cardiac images can be analyzed during vascular analysis to generate 3D models and determine the vascular characteristics of the cardiac vasculature depicted in the cardiac images. These cardiac images may depict the portion from different viewpoints (e.g., by using a c-arm), such that a system may generate a three-dimensional model of the portion.SUMMARY

[0005] The disclosure generally contemplates systems and methods for providing guidance to a medical professional during coronary physiology assessments, for example to more accurately characterize a patient’s cardiac health options.

[0006] In some aspects, the techniques described herein relate to a method implemented by a system of one or more processors including: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature including one or more vessels; displaying, via a user interface, at least one medical image of the plurality of medical image data; receiving information annotating, via the user interface, vessel markings for at least one indicator of the one or more vessels of the subject of the displayed at least one medical image; and concurrently displaying, via the user interface, at least one prompt on the at least one medical image by the system, wherein each of the at least one prompt corresponds to user instructions for annotating a specific indicator of the at least one indicators.

[0007] In some aspects, the techniques described herein relate to a method, wherein each of the at least one prompt is positioned on the user interface proximate to corresponding indicator.

[0008] In some aspects, the techniques described herein relate to a method, wherein the specific indicator is visible when displaying the corresponding prompt on the user interface.

[0009] In some aspects, the techniques described herein relate to a method, wherein the at least one prompt is re-positioned responsive to user input via the user interface.

[0010] In some aspects, the techniques described herein relate to a method, further including selecting a target vessel, wherein the one or more vessels includes the target vesseland the at least one indicator includes at least one of an ostium, a lesion, a selected target vessel, or at least one major side branch, and the at least one prompt includes at least one of an ostium prompt, a lesion prompt, a selected vessel prompt, or a major side branch prompt corresponding to each of the at least one indicator.

[0011] In some aspects, the techniques described herein relate to a method, wherein the ostium prompt corresponds to at least one of user instructions including a suggestion to adjust the annotation of the ostium to a correct location or verification that the annotation of the ostium is in the correct location and wherein the suggestion for the correct location of the ostium is associated with one or more diameters of the target vessel.

[0012] In some aspects, the techniques described herein relate to a method, wherein the major side branch prompt corresponds to user instructions to annotate the at least one major side branch in a first medical image of the at least one medical images, and wherein the user instructions to annotate are associated with a missing major side branch vessel annotated in a second medical image of the at least one medical images and not in the first medical image, wherein the missing major side branch vessel is well-demonstrated in the first medical image.

[0013] In some aspects, the techniques described herein relate to a method, wherein receiving information annotating the vessel markings for the at least one indicator of the one or more vessels includes: receiving user input that annotates a lesion marker on the at least one medical image; generating the vessel markings for each of the at least one indicator by the system in the at least one medical image; and at least one of: verifying accuracy of the generated vessel markings, or receiving user inputs to adjust at least one of the generated vessel markings for each of the at least one indicator.

[0014] In some aspects, the techniques described herein relate to a method, wherein displaying, via the user interface, at least one prompt includes displaying the at least one prompt in a pre-determined order, and wherein the method further includes displaying a next prompt in response to a corresponding user input received by the user interface, the corresponding user input including the user selecting a next button or adjusting the generated annotation of the specific indicator corresponding to the displayed prompt.

[0015] In some aspects, the techniques described herein relate to a method, wherein the method further includes displaying a previous prompt in response to a return user input received by the user interface.

[0016] In some aspects, the techniques described herein relate to a method, wherein the method further includes, in response to a non-corresponding input: displaying, via the user interface, the prompt corresponding to a non-corresponding indicator, the non-corresponding input including adjusting, via the user interface, the generated annotation of the noncorresponding indicator, the non-corresponding indicator including the at least one indicators that do not correspond to the displayed prompt.

[0017] In some aspects, the techniques described herein relate to a method, wherein the user instructions includes at least one of: text, images, or animation.

[0018] In some aspects, the techniques described herein relate to a method, wherein the user instructions includes an example of an adjusted vessel marking.

[0019] In some aspects, the techniques described herein relate to a method, further including: generating a three-dimensional model depicting the one or more vessels using the annotated at least one medical image; displaying, via the user interface, the three-dimensional model.

[0020] In some aspects, the techniques described herein relate to a method, further including: determining an index indicative of vascular function based on the three-dimensional model.

[0021] In some aspects, the techniques described herein relate to a method, further including receiving, via the user interface, at least one additional prompt by the system for selecting a frame, wherein each of the medical image data in the plurality of medical image data includes a medical video including a plurality of frames.

[0022] In some aspects, the techniques described herein relate to a method, wherein the at least one additional prompt for selecting the frame corresponds to user instructions for selecting an end diastolic frame, wherein the end diastolic frame depicts the one or more vessels during an end diastolic phase, and wherein the at least one additional prompt is received in response to determining the plurality of frames include a cycle of systolic and end diastolic frames.

[0023] In some aspects, the techniques described herein relate to a method, further including receiving, via the user interface, at least one additional prompt by the system for selecting, via the user interface, a target vessel of the vasculature of the subject.

[0024] In some aspects, the techniques described herein relate to a method, further including receiving, via the user interface, at least one prompt by the system for selecting, via the user interface, a subset of the plurality of medical image data.

[0025] In some aspects, the techniques described herein relate to a method, further including: receiving user input, via the user interface, to either annotate at least one of the at least one indicators or confirm that the at least one indicators are correctly annotated; and generating a model of at least one of the vessels based on the displayed at least one medical image and at least one of the information or the user input.

[0026] In some aspects, the techniques described herein relate to a method, wherein at least one of the prompt includes a smart prompt that corresponds to user instructions for a suggestion to annotate an indicator of the at least one indicators.

[0027] In some aspects, the techniques described herein relate to a method, further including: determining a suggested annotation associated with a correct annotation for the indicator based on one or more analyses of the at least one medical image, wherein the suggestion corresponds with the suggested annotation; receiving user input, via the user interface, to either annotate or verify that the indicator has the correct annotation on the displayed at least one medical image; and generating a model of at least one of the one or more vessel based on the displayed at least one medical image, wherein the indicator having the correct annotation in the at least one medical image is usable to generate the model.

[0028] In some aspects, the techniques described herein relate to a method, further including: determining a suggestion associated with annotating an indicator of the at least one indicators, and replacing a corresponding prompt associated with the indicator with a smart prompt that corresponds to user instructions based on the suggestion.

[0029] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0030] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0031] In some aspects, the techniques described herein relate to a method implemented by a system of one or more processors including: displaying, via a user interface, at least onemedical image of a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature including one or more vessels; displaying, via the user interface, over the at least one medical image, vessel markings for at least one indicator of the one or more vessels of the subject; and concurrently displaying, via the user interface, at least one prompt over the at least one medical image by the system, wherein each of the at least one prompt corresponds to instructions for a user to annotate a specific indicator of the at least one indicator.

[0032] In some aspects, the techniques described herein relate to a method wherein displaying the vessel markings for the at least one indicator of the one or more vessels includes: presenting a lesion marker at a selected position on the at least one medical image in response to receiving user input for the lesion marker; displaying the vessel markings for each of the at least one indicator in the at least one medical image, the vessel markings generated by the system; and displaying a next prompt of the at least one prompts in response to at least one of: receiving user input verifying accuracy of the generated vessel marking for the specific indicator corresponding to a currently displayed prompt, or receiving user input to adjust the generated vessel marking for the specific indicator corresponding to the currently displayed prompt.

[0033] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0034] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0035] In some aspects, the techniques described herein relate to a method of annotating one or more vessels of a medical image for generating a model of the one or more vessels, including: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature including the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical image data; receiving information annotating, via the user interface, each of the at least two medical images with at least one marker selected from: (i) an ostial marker corresponding with an ostium of the one or more vessels, (ii) a set of major side branch markers corresponding with a set of major sidebranches of the one or more vessels, (iii) a lesion marker associated with a lesion of the one or more vessels, the lesion marker associated with a boundary near the lesion, and (iv) a selected vessel marker corresponding with a selected vessel of the one or more vessels, the selected vessel markers associated with a boundary of the selected vessel; displaying, via the user interface, at least one prompt to adjust the at least one marker for a correct annotation or to verify that the at least one marker has the correct annotation; wherein the at least one marker having the correct annotation for each of the at least two medical images is usable to generate the model of the one or more vessels.

[0036] In some aspects, the techniques described herein relate to a method, further including, in response to receiving user input associated with the prompt to adjust or verify a first marker of the at least one marker for a first medical image of the at least two medical images: removing a first prompt of the at least one prompt from displaying, the first prompt corresponding to the first marker; and displaying, via the user interface, a second prompt corresponding to user instructions to adjust a second marker for a correct annotation in a corresponding second medical image of the at least one medical images or to verify that the second marker has the correct annotation in the second medical image, the second prompt being different from the first prompt, and the second marker being a same type as the first marker.

[0037] In some aspects, the techniques described herein relate to a method, further including, in response to receiving user input associated with the prompt to adjust or verify a first marker of the at least one marker for a first medical image of the at least two medical images: removing a first prompt of the at least one prompt from displaying, the first prompt corresponding to the first marker; and displaying, via the user interface, a second prompt corresponding to user instructions to adjust a second marker on the first medical image to a correct annotation or to verify that the second marker is correctly annotated on the first medical image, the second prompt being different from the first prompt, and the second marker being different from the first marker.

[0038] In some aspects, the techniques described herein relate to a method, wherein at least one of the at least one prompt is a smart prompt generated based on one or more analyses of at least one medical image of the at least two medical images

[0039] In some aspects, the techniques described herein relate to a method, further including, in response to receiving user input associated with the prompt to adjust or verify a first marker of the at least one marker for a first medical image of the at least two medical images: removing the smart prompt from displaying, the smart prompt corresponding to the first marker; and displaying, via the user interface, a non-smart prompt corresponding to user instructions to verify that the first marker is correctly annotated on the first medical image, the non-smart prompt being different from the smart prompt and corresponding to the first marker.

[0040] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0041] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0042] In some aspects, the techniques described herein relate to a method implemented by a system of one or more processors including: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature including one or more vessels; receiving information annotating one or more indicators for the one or more vessels in at least one medical image of the plurality of medical image data; generating a prompt set including a prompt for each of the one or more indicators to verify that each of the one or more indicators has a correct annotation; determining a suggestion associated with the correct annotation of at least one indicator of the one or more indicators based on at least one of the information annotating the one or more indicators and one or more analyses of the at least one medical image; generating a smart prompt based on the suggestion corresponding to user instructions to annotate the at least one indicator for the correct annotation; and replacing a corresponding prompt of the one or more prompt with the smart prompt in the prompt set; wherein the correct annotation of each of the one or more indicators on the at least one medical image is usable to generate a model of the one or more vessels.

[0043] In some aspects, the techniques described herein relate to a method, further including receiving user input associated with the suggestion, the user input including either a user annotation of the at least one indicator associated with the suggestion or a removal of the suggestion.

[0044] In some aspects, the techniques described herein relate to a method, further including replacing the smart prompt with the corresponding prompt in the prompt set in response to receiving the user input.

[0045] In some aspects, the techniques described herein relate to a method, wherein determining a suggestion includes determining a score associated with the annotation of the at least one of the one or more indicators and comparing the score with a suggestion threshold.

[0046] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0047] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0048] In some aspects, the techniques described herein relate to a method implemented by a system of one or more processors including: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature including one or more vessels; receiving information annotating one or more indicators for the one or more vessels in at least one medical image of the plurality of medical image data; determining at least one suggestion associated with a correct annotation of at least one indicator of the one or more indicators based on at least one of the information annotating the one or more indicators and one or more analyses of the at least one medical image; generating a sequence of banners including a smart banner for each of the at least one suggestion, each of the smart banners corresponding to user instructions to annotate the at least one indicator for a correct annotation or to verify that the at least one indicator has a correct annotation; displaying a first smart banner of the sequence of banners; and replacing the first smart banner and displaying a second smart banner of the sequence of banners in response to a corresponding input; wherein the correct annotation of each of the one or more indicators on the at least one medical image is usable to generate a model of the one or more vessels.

[0049] In some aspects, the techniques described herein relate to a method, wherein the corresponding input includes one or more of: a user annotation of the one or more vessels associated with a particular suggestion of the at least one suggestion, a user input requesting another banner of the sequence of banners, or a removal of the particular suggestion.

[0050] In some aspects, the techniques described herein relate to a method, further including removing the first smart banner from the sequence of banners in response to the corresponding input received including the user annotation of the one or more vessels associated with the particular suggestion or the removal of the particular suggestion.

[0051] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0052] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0053] In some aspects, the techniques described herein relate to a method of annotating one or more vessels of a medical image for generating a model of the one or more vessels, including: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature including the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical image data each including an ostium of the one or more vessels; receiving information annotating, via the user interface, each of the at least two medical images with an ostial marker corresponding with the ostium; and displaying, via the user interface, an ostium prompt corresponding to user instructions to adjust at least one of the ostial markers to a correct location associated with the ostium in the corresponding medical image of the at least two medical images or to verify that the at least one of the ostial marker is in the correct location; wherein the ostial markers being in the correct location for each of the at least two medical images is usable to generate the model of the one or more vessels.

[0054] In some aspects, the techniques described herein relate to a method, wherein the ostium prompt is a smart prompt generated based on one or more analyses of the corresponding medical image of the at least two medical images.

[0055] In some aspects, the techniques described herein relate to a method, wherein the smart prompt is based on a comparison of vessel diameters of at least a portion of the one or more vessels to a threshold value associated with vessel diameter.

[0056] In some aspects, the techniques described herein relate to a method, wherein the threshold value is based on a metric of central tendency of the vessel diameters of at least one of the one or more vessels.

[0057] In some aspects, the techniques described herein relate to a method, wherein the portion of the one or more vessels includes a percent of a length of the selected vessel closest to the ostium, the length being from the ostium to a distal end of the selected vessel.

[0058] In some aspects, the techniques described herein relate to a method, wherein the percent is between 5-25%.

[0059] In some aspects, the techniques described herein relate to a method, wherein the threshold value is based on the vessel diameters of a remainder of the selected vessel not included in the portion of the one or more vessels.

[0060] In some aspects, the techniques described herein relate to a method, wherein the threshold value is based on a metric of central tendency of the remainder of the selected vessel.

[0061] In some aspects, the techniques described herein relate to a method, further including measuring the vessel diameters of the at least the portion of the one or more vessels.

[0062] In some aspects, the techniques described herein relate to a method, wherein a first number of measurements of vessel diameters are measured along a first portion of the one or more vessels, the first number of measurements used in the comparison to the threshold value.

[0063] In some aspects, the techniques described herein relate to a method, further including determining a directionality associated with the at least the portion of the one or more vessels, and wherein measuring the vessel diameters includes compensating for the directionality of the one or more vessels.

[0064] In some aspects, the techniques described herein relate to a method, wherein the threshold value includes a threshold rate based on a rate of change between the vessel diameters.

[0065] In some aspects, the techniques described herein relate to a method, wherein the threshold rate is associated with at least one of a quick broadening or a quick narrowing.

[0066] In some aspects, the techniques described herein relate to a method, wherein the threshold rate is associated with at least one of a slow broadening or a slow narrowing.

[0067] In some aspects, the techniques described herein relate to a method, wherein the comparison includes comparing a rate of change between the vessel diameters of a portion of a selected vessel of the one or more vessels closest to the ostium to the threshold rate.

[0068] In some aspects, the techniques described herein relate to a method, wherein the user instructions corresponding to the ostium prompt includes a suggestion associated with the correct location based on the vessel diameters of at least a portion of the one or more vessels.

[0069] In some aspects, the techniques described herein relate to a method, further including, in response to receiving the user instructions to adjust or verify a first ostial marker of a first medical image of the at least two medical images: removing a first ostium prompt from displaying; and displaying, via the user interface, a second ostium prompt corresponding to user instructions to adjust a second ostial marker different from the first ostial marker to a correct location associated with the ostium in a corresponding second medical image of the at least one medical images or to verify that the second ostial marker is in the correct location.

[0070] In some aspects, the techniques described herein relate to a method, further including, in response to receiving the user instructions to adjust or verify the ostial marker for a first medical image of the at least two medical images: removing the ostium prompt from displaying; and displaying, via the user interface, an indicator prompt different from the ostium prompt corresponding to user instructions to adjust an indictor marker different from the ostium marker on the first medical image for a correct annotation or to verify that the indicator marker is correctly annotated.

[0071] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0072] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0073] In some aspects, the techniques described herein relate to a method of annotating one or more vessels of a medical image for generating a model of the one or more vessels, including: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature including the one or more vessels; displaying, via a user interface, a first medical image and a second medical image of the plurality of medical imagedata, the first medical image including a first set of major side branches of the one or more vessels and the second medical image including a second set of major side branches; receiving information annotating, via the user interface, the first medical image with a first set of major side branch markers corresponding with the first set of major side branches and annotating the second medical image with a second set of major side branch markers corresponding with the second set of major side branches; and displaying, via the user interface, a major side branch prompt corresponding to user instructions to adjust the first set of major side branch markers to include requisite markings associated with the first set and second set of major side branches or to verify that the first set of major side branch markers are annotated to include the requisite markings; wherein the first set and the second of major side branch markers being annotated to include the requisite markings in the first medical image and the second medical image is usable to generate the model of the one or more vessels.

[0074] In some aspects, the techniques described herein relate to a method, wherein the major side branch prompt is a smart prompt generated based on one or more analyses of at least one of the first medical image and the second medical image.

[0075] In some aspects, the techniques described herein relate to a method, wherein the smart prompt is based on a comparison of the first set of major side branch markers with the second set of major side branch markers.

[0076] In some aspects, the techniques described herein relate to a method, wherein the requisite markings are associated with a number of major side branches, the comparison is based on a first number of major side branches annotated by the first set of major side branch markers and a second number of major side branches annotated by the second set of major side branch markers, smart prompt being generated if the first number is less than the second number.

[0077] In some aspects, the techniques described herein relate to a method, wherein the smart prompt is based on a comparison of a corresponding geometry of each of the major side branches annotated by the first set of major side branch markers and the second set of major side branch markers, wherein a type of major side branch is determined based on the corresponding geometry.

[0078] In some aspects, the techniques described herein relate to a method, wherein the comparison between a first set of types of major side branches determined for the first set ofmajor side branch markers and a second set of types determined for the second set of major side branch markers, the smart prompt generated if the first set of types and the second set of types differ.

[0079] In some aspects, the techniques described herein relate to a method, wherein the smart prompt corresponds to user instructions to annotate at least one major side branch associated with the second set of major side branch markers and not the first set of major side branch markers.

[0080] In some aspects, the techniques described herein relate to a method, wherein the second set of major side branch markers includes the annotation of a bifurcated major side branch in the second set of major side branches that is not included in the first set of major side branch markers, and the smart prompt corresponds to user instruction to adjust the first set of major side branch markers to include the requisite markings, the requisite markings including the bifurcated major side branch in the first medical image.

[0081] In some aspects, the techniques described herein relate to a method, wherein a bifurcation of the bifurcated major side branch is near a lesion depicted in the one or more vessels.

[0082] In some aspects, the techniques described herein relate to a method, wherein the information associated with annotating the first medical image includes a confidence score for each of the major side branches in the first set of major side branches, wherein a major side branch marker is annotated on the major side branch when the associated confidence score is above a confidence threshold.

[0083] In some aspects, the techniques described herein relate to a method, wherein the at least one major side branch is well-demonstrated in the first medical image.

[0084] In some aspects, the techniques described herein relate to a method, wherein the at least one major side branch is well-demonstrated when the information detects the at least one major side branch but the associated confidence score is below the confidence threshold.

[0085] In some aspects, the techniques described herein relate to a method, wherein the confidence score is based on a path associated with each major side branch of the first set of major side branches, the path determined based on a regression model.

[0086] In some aspects, the techniques described herein relate to a method, wherein the confidence score is based on a mask associated with each major side branch of the first set of major side branches, the mask determined based on a segmentation model.

[0087] In some aspects, the techniques described herein relate to a method, further including, in response to receiving the user instructions to adjust or verify the first set of major side branch markers of the first medical image: removing a first major side branch prompt from displaying; and displaying, via the user interface, a second major side branch prompt corresponding to user instructions to adjust the second set of major side branch markers to include the requisite markings or to verify that the second set of major side branch markers are annotated to include the requisite markings.

[0088] In some aspects, the techniques described herein relate to a method, further including, in response to receiving the user instructions to adjust or verify the first set of major side branch markers for the first medical image: removing the major side branch prompt from displaying; and displaying, via the user interface, an indicator prompt different from the major side branch prompt corresponding to user instructions to adjust an indictor marker different from the first set of major side branch markers on the first medical image for a correct annotation or to verify that the indicator marker is correctly annotated.

[0089] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0090] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0091] In some aspects, the techniques described herein relate to a method of annotating one or more vessels of a medical image for generating a model of the one or more vessels, including: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature including the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical image data each including a lesion of the one or more vessels; receiving information annotating, via the user interface, each of the at least two medical images with a lesion marker corresponding with the lesion; and displaying, via the user interface, a lesion prompt corresponding to user instructions to adjustat least one of the lesion markers to include correct contours associated with the lesion in the corresponding medical image of the at least two medical images or to verify that the at least one of the lesion marker has the correct contours; wherein the lesion markers having the correct contours for each of the at least two medical images is usable to generate the model of the one or more vessels.

[0092] In some aspects, the techniques described herein relate to a method, wherein the lesion prompt is a smart prompt generated based on one or more analyses of the corresponding medical image of the at least two medical images.

[0093] In some aspects, the techniques described herein relate to a method, further including, in response to receiving the user instructions to adjust or verify a first lesion marker of a first medical image of the at least two medical images: removing a first lesion prompt from displaying; and displaying, via the user interface, a second lesion prompt corresponding to user instructions to adjust a second lesion marker different from the first lesion marker to include correct contours associated with the lesion in a corresponding second medical image of the at least one medical images or to verify that the second lesion marker has the correct contours.

[0094] In some aspects, the techniques described herein relate to a method, further including, in response to receiving the user instructions to adjust or verify the lesion marker for a first medical image of the at least two medical images: removing the lesion prompt from displaying; and displaying, via the user interface, an indicator prompt different from the lesion prompt corresponding to user instructions to adjust an indictor marker different from the lesion marker on the first medical image for a correct annotation or to verify that the indicator marker is correctly annotated.

[0095] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0096] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0097] In some aspects, the techniques described herein relate to a method of annotating one or more vessels of a medical image for a model of the one or more vessels, including: accessing a plurality of medical image data depicting a portion of a vasculature of a subject,the portion of the vasculature including the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical image data each including a selected vessel of the one or more vessels; receiving information annotating, via the user interface, each of the at least two medical images with a selected vessel marker corresponding with the selected vessel; and displaying, via the user interface, a selected vessel prompt corresponding to user instructions to adjust at least one of the selected vessel marker to include correct contours associated with the selected vessel in the corresponding medical image of the at least two medical images or to verify that the at least one of the selected vessel marker has the correct contours; wherein the selected vessel markers having the correct contours for each of the at least two medical images is usable to generate the model of the one or more vessels.

[0098] In some aspects, the techniques described herein relate to a method, wherein the selected vessel prompt is a smart prompt generated based on one or more analyses of the corresponding medical image of the at least two medical images.

[0099] In some aspects, the techniques described herein relate to a method, further including, in response to receiving the user instructions to adjust or verify a first selected vessel marker of a first medical image of the at least two medical images: removing a first selected vessel prompt from displaying; and displaying, via the user interface, a second selected vessel prompt corresponding to user instructions to adjust a second selected vessel marker different from the first selected vessel marker to include correct contours associated with the selected vessel in a corresponding second medical image of the at least one medical images or to verify that the second selected vessel marker has the correct contours.

[0100] In some aspects, the techniques described herein relate to a method, further including, in response to receiving the user instructions to adjust or verify the selected vessel marker for a first medical image of the at least two medical images: removing the selected vessel prompt from displaying; and displaying, via the user interface, an indicator prompt different from the selected vessel prompt corresponding to user instructions to adjust an indictor marker different from the selected marker on the first medical image for a correct annotation or to verify that the indicator marker is correctly annotated.

[0101] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that whenexecuted by the one or more processors, cause the one or more processors to perform the method.

[0102] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.

[0103] In some aspects, the techniques described herein relate to a method of annotating one or more vessels of a medical image for generating a model associated with the one or more vessels, the method including: accessing a plurality of medical images depicting a portion of a vasculature of a subject, the portion of the vasculature including the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical images; obtaining information annotating at least one of the at least two medical images with at least one marker corresponding to one or more vessel features, the at least one marker including at least one of: (i) an ostial marker corresponding with an ostium of the one or more vessels, (ii) one or more major side branch markers corresponding with one or more major side branches of the one or more vessels, (iii) a lesion marker corresponding to a lesion of the one or more vessels, the lesion marker associated with a boundary of the one or more vessels near the lesion, and (iv) a selected vessel marker corresponding with a selected vessel of the one or more vessels, the selected vessel markers associated with a boundary of the selected vessel, wherein the one or more vessel features includes at least one of the ostium, the one or more major side branches, the lesion, and the selected vessel; generating, for display on a user interface, at least one prompt configured to cause adjustment of the at least one marker, the adjustment being configured to correct the annotation or verify the at least one marker; wherein the adjustment is configured to enhance generation of the model.

[0104] In some aspects, the techniques described herein relate to a method, further including, in response to receiving user input associated with a first prompt of the at least one prompt in a first medical image: removing the first prompt of the at least one prompt from displaying, the first prompt corresponding to a first marker; displaying, via the user interface: a second prompt associated with a second marker in the first medical image, the second marker associated with a different vessel feature from the first marker if user input associated with each of the at least one prompt in the first medical image has not been received; or a third prompt associated with the first marker in a corresponding second medical image if user input associated with each of the at least one prompt in the first medical image has been received.

[0105] In some aspects, the techniques described herein relate to a method, wherein at least one of the at least one prompt is a smart prompt generated based on one or more analyses of at least one medical image of the at least two medical images, the smart prompt corresponding to user instructions to adjust the at least one marker for the correct annotation.

[0106] In some aspects, the techniques described herein relate to a method, further including, in response to not receiving user input corresponding to the user instructions of the smart prompt, preventing generation of the model of the one or more vessels.

[0107] In some aspects, the techniques described herein relate to a method, wherein the smart prompt is generated based on one or more analyses of the at least one marker annotating at least one of the at least two medical images.

[0108] In some aspects, the techniques described herein relate to a method, wherein the one or more analyses include comparing the at least one vessel features between the at least two medical images.

[0109] In some aspects, the techniques described herein relate to a method, wherein the one or more vessel features includes at least the one or more major side branches, wherein the one or more analyses includes comparing at least one of a type or a number of the vessel markers for the one or more major side branches between the at least two medical images, wherein the smart prompt is generated if at least one of the type or the number of the vessel markers is different between the at least two medical images.

[0110] In some aspects, the techniques described herein relate to a method, wherein the one or more vessel features includes at least the ostium, wherein the one or more analyses includes determining diameters of the one or more vessels proximal to the ostium, wherein the smart prompt is generated if a change between the diameters is above a threshold.

[0111] In some aspects, the techniques described herein relate to a method, wherein the one or more analyses to generate the smart prompt occurs simultaneously with generating the information annotating at least one of the at least two medical images with the at least one marker.

[0112] In some aspects, the techniques described herein relate to a method, further including determining a confidence score associated with a particular vessel feature of the one or more vessel features based on the at least one medical image, wherein: the one or more analyses occurs in response to determining that the confidence score is below an annotationthreshold, or the information associated with annotating the particular vessel feature with the at least one marker is received in response to determining that the confidence score is above the annotation threshold.

[0113] In some aspects, the techniques described herein relate to a method, wherein the confidence score is based on contrast of the particular vessel feature determined from the at least one medical image.

[0114] In some aspects, the techniques described herein relate to a method, wherein the one or more analyses includes: comparing the confidence score to a suggestion threshold; and in response to the confidence score being above a suggestion threshold, generating the smart prompt.

[0115] In some aspects, the techniques described herein relate to a method, further including, in response to receiving user input: validating whether the user input is adjusting or verifying a first marker of the at least one marker for a first medical image of the at least two medical images to the correct annotation, the correct annotation associated with the user instructions of the smart prompt; in response to the validation of the user input, removing the smart prompt from displaying, the smart prompt corresponding to the first marker.

[0116] In some aspects, the techniques described herein relate to a method, further including, in response to removing the smart prompt from displaying, displaying, via the user interface, a non-smart prompt corresponding to user instructions to verify that the first marker is correctly annotated on the first medical image, the non-smart prompt being different from the smart prompt and corresponding to the first marker.

[0117] In some aspects, the techniques described herein relate to a method, further including, in response to receiving user input associated a first smart prompt corresponding to a first marker for a first medical image: removing the first smart prompt from displaying; and displaying, via the user interface: a second smart prompt corresponding to the first marker for the first medical image if user input associated with each of the smart prompt for the first medical image has not been received, the second smart prompt corresponding to different user instructions to adjust the first marker for the correct annotation than the first smart prompt; and a third smart prompt corresponding to a second marker of the at least one marker for the first medical image if a user input associated with each of the smart prompt for the first medical image has been received.

[0118] In some aspects, the techniques described herein relate to a method, wherein each of the at least one prompt is positioned on the user interface proximate to corresponding vessel feature.

[0119] In some aspects, the techniques described herein relate to a method, wherein the at least one prompts include at least one banner.

[0120] In some aspects, the techniques described herein relate to a method, wherein the ostial marker corresponds with the ostium of the selected vessels.

[0121] In some aspects, the techniques described herein relate to a method, wherein obtaining information annotating the at least two medical images with at least one marker corresponding to one or more vessel features includes: receiving user input annotating a location of the lesion on at least one of the at least two medical image; and at least one of: automatically generating the at least one marker for the one or more vessel features in the at least one medical image; or receiving user inputs to adjust at least one of the generated vessel markers for each of the at least one feature.

[0122] In some aspects, the techniques described herein relate to a method, wherein the at least one prompt includes at least one of text, images, or animation associated with the at least one marker having the correct annotation.

[0123] In some aspects, the techniques described herein relate to a method, wherein the at least one marker having the correct annotation for at least one of the at least two medical images increases an accuracy of a 3D model of the one or more vessels generated from the at least two medical images.

[0124] In some aspects, the techniques described herein relate to a system including one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method.

[0125] In some aspects, the techniques described herein relate to non-transitory computer storage media storing instructions that when executed by a system.BRIEF DESCRIPTION OF THE DRAWINGS

[0126] Throughout the drawings, reference numbers are re-used to indicate correspondence between referenced elements. The drawings are provided to illustrate embodiments of the subject matter described herein and not to limit the scope thereof.

[0127] Figure 1 is a block diagram of an example cardiac image analysis system for selecting and annotating cardiac images to generate a three-dimensional model.

[0128] Figure 2 is a block diagram illustrating an annotation engine and user assistance engine configured to generate annotated cardiac images.

[0129] Figure 3A is a flow diagram of an example process for generating annotated images using one or more prompts.

[0130] Figure 3B is a flow diagram of an example process for verifying annotations based on one or more prompts.

[0131] Figure 4A illustrates an example user interface displaying example annotations with user assistance configured to be off.

[0132] Figure 4B illustrates the example user interface displaying an example annotation of a selected image with user assistance configured to be on.

[0133] Figure 5A illustrates an example ostium prompt.

[0134] Figure 5B illustrates an example lesion prompt.

[0135] Figure 5C illustrates an example selected vessel prompt.

[0136] Figure 5D illustrates an example major side branch prompt is displayed.

[0137] Figure 6A is a flow diagram of an example process for generating smart prompts.

[0138] Figure 6B is a flow diagram of an example process for generating smart banners when user assistance is configured to be off.

[0139] Figure 7A illustrates an example of an ostium smart prompt.

[0140] Figure 7B illustrates an example of a major side branch smart prompt.

[0141] Figure 8A illustrates an example of smart banners for the ostium when user assistance is configured to be off.

[0142] Figure 8B illustrates an example of smart banners for the major side branch when user assistance is configured to be off.

[0143] Figure 8C illustrates an example of smart banners for the selected vessel when user assistance is configured to be off.DETAILED DESCRIPTIONOverview

[0144] This specification describes techniques to provide guidance to a medical professional during coronary physiology assessments, for example to more accurately characterize a patient’s cardiac health options. Cardiac images can be analyzed during vascular analysis to generate 3D models and determine the vascular characteristics of the cardiac vasculature depicted in the cardiac images. Further descriptions related to automatic selection of cardiac images, vascular computer modeling, segmentation and analysis, and related systems and methods are included in U.S. Patent Pub. 2023 / 0252632, which is hereby incorporated herein by reference in its entirety.

[0145] The vascular characteristics can include indices indicative of vascular function or index drops (e.g., reductions in FFR value) along a selected vessel that can illustrate the health of the selected vessel and indicate the location of lesions along the vessel. An example index may include a fractional flow reserve (FFR) value or a vascular state score. A medical professional may leverage the indices described herein to identify one or more lesions which are to be analyzed. A lesion, as may be appreciated, may be associated with a narrowing or constriction of a cardiac vessel. Description related to determining or calculating an index indicative of vascular function is included in U.S. Patent No. 10,595,807 and incorporated herein by reference in its entirety. Description related to determining the vascular state score is included in U.S. Patent No. 10,943,233, which is hereby incorporated herein by reference in its entirety.

[0146] A user (e.g., a medical professional) may need to provide input regarding the cardiac images prior to their use in the vascular analysis. For example, to output an accurate representation of vascular health, the system may require that a user properly annotate the location of vessel indicators (or “vessel features”), including the vessel ostium, the boundaries of the lesion, the boundaries of the selected vessel, and that each major side branch of the selected vessel, for each cardiac image selected by the user for the vascular analysis. The boundaries can include tracings or contours. A user unfamiliar with how to provide said input may provide incorrect information that upon vascular analysis, may provide less accurate results on the health of the cardiac vasculature. While training can be provided to users on how to properly provide input during coronary physiology assessments, users may not fullycomprehend the training during the session or may need to receive refreshers to optimize the assessment.

[0147] Instead, user training or assistance can be made readily available so that a user can practice with the system. Alternatively, user training or assistance can also be available during the implementation of each coronary physiology assessment during clinical use. The user training or assistance can be provided through alerts, such as prompts, which are displayed to the user and are integrated into the coronary physiology assessment system. These alerts, which may be referred to as prompts or banners but are not limited to only prompts or banners, can guide a user on how to operate the coronary physiology assessment system in a similar manner as provided during training sessions or can even include additional guidance than the training sessions. Thus, user training or assistance could not only reduce expenses and time related to training outside of operating the assessment but increase the accuracy of coronary physiology assessments by both new and experienced users. Accordingly, having the annotations in a correct location or having a correct boundary (e.g., where the annotation matches the vessel feature in the medical image) can increase the accuracy of the model generated from the medical images. Additionally, the training can be tailored to the cardiac images and provide suggestions on how and where to annotate to provide impactful effects on the analysis. The user can also have the option to turn off the assistance as needed, which can still include more limited assistance to the user.

[0148] “QCA” is not intended to be limiting and may be used to refer to any other minimally invasive coronary physiology assessment, such as 2D radiography, 3D quantitative assessments, and so on.

[0149] ‘Ischemia,” “stenosis,” and / or “coronary stenosis” are not intended to be limiting and may be interchanged or refer to any other condition related to the narrowing of the vessels that may be treated through revascularization. “Lesions” refer to the portion of the cardiac vasculature where the vessel is narrower and is not intended to be limited to relating to ischemia or coronary stenosis, but to any CAD.

[0150] In some embodiments, additionally or alternatively, the vasculature may be of another organ, for example, a kidney, a retina, and / or a brain. It should be understood, where cardiac vasculature is described in particular, that implicit reference is also made to embodiments relating to the vasculature of another organ.Block Diagram

[0151] Figure 1 is a block diagram of an example cardiac image analysis system 100 for selecting and annotating cardiac images 110 to generate a three-dimensional model 180. The cardiac image analysis system 100 may represent a system of one or more processors, such as a user device, a computer, a back-end server system, and so on. As described above, the system 100 may receive cardiac images 110 (e.g., angiograms) for analysis. The system 100 may additionally present a user interface 101 for interaction with an end-user (e.g., a medical professional) via user input 160. The system 100 may, in some embodiments, present the user interface 101 via a display of the system. The system 100 may also cause presentation of the user interface 101 via a display of a different system or user device (e.g., the user interface 101 may represent a front-end of a web application). The user interface 101 may be used by the end-user to prepare cardiac images 110 for further analysis by the system 100.

[0152] During image and vessel selection 120, the medical professional may provide input 160 specifying target vessels of the heart, such as the left coronary artery (LCA), left anterior descending artery (LAD), right coronary artery (RCA), etc. The system 100 may then provide a subset of the cardiac images 110 that depict a vessel of the cardiac vasculature of choice, and the medical professional may input 160 a selection of which of these cardiac images to annotate and analyze. In some instances, the selected images 130 may include three of the cardiac images 110. As described above, and as incorporated herein, the selected images 130 may depict the selected vessel from different viewpoints. Information from these viewpoints may be combined to form the three-dimensional model 180. Each medical image may represent an image from a multitude of images which were taken at a particular viewpoint.

[0153] The medical professional may also need to select values associated with the cardiac vasculature, such as mean aortic pressure. Within each of the selected images 130, the medical professional may identify a primary lesion, or set of lesions, on the selected vessel which is depicted in each accessed cardiac image. User input 160 may be provided to each cardiac image that indicates a portion of the patient’s heart that has a lesion. For example, the medical professional may indicate the position of the lesion via user input 160 to control a lesion marker, such as a plus sign inside of the circle. Based on this user input, the annotation engine 140 may then identify and annotate markers for different portions of the patient’s cardiacvasculature, or vessel indicators (or “vessel features”), that are associated with the lesion, such as the vessel ostium connecting the selected vessel to the aorta, the boundaries of the vessel near the lesion, the boundaries of the selected vessel, and the markings of and / or boundaries of the major side branches of the selected vessel. As mentioned above, the boundaries can include contours or tracings.

[0154] The medical professional may review the annotation and may prefer to adjust the vessels markers identified by the annotation engine 140 via user input 160 to the annotation engine 140. Thus, annotation 140 includes automatic annotation (or “projections”) and either verification or adjustments of the annotations via user input 160. For example, the medical professional may provide user input 160 to adjust the vessel markers for the vessel indicators (or “vessel features”), such as by extending and / or adding to the identified vessels (including the selected vessel and the major side branches), adjusting the vessel boundaries, and adjusting the selected vessel ostium in each of the selected images 130. For example, the medical professional can extend an identified vessel, redraw or reshape the boundary annotations of the vessels, and / or move the ostium marker or other portion of the vessel via dragging a graphical element representing the identified vessel. Thus, adjusting the vessels may include adjusting the length or other geometric characteristics of a vessel (e.g., increasing or decreasing a vessel). Adjusting the vessels may also include, for example as described, adding an additional vessel or vessel segment as being relevant to a lesion. Adjusting the vessels may also include, for example, removing a vessel or vessel segment as not being relevant. For example, the system may have misclassified a vessel, such as a major side branch, as having blood flow that goes into or extends from a lesion. In some instances, the annotation engine 140 may only include annotation via user input 160 by the medical professional. During annotation 140, the user can adjust the lesion marker to restart the automatic annotation. Alternatively, the user may need to restart annotation for all of the selected images 130 to adjust the lesion marker on any of the selected images 130.

[0155] During the annotation 140 process, the user assistance engine 150 can be turned on and presented via the user interface 101. During the annotation process, the user assistance engine 150 can provide prompts associated with the types of vessel annotation generated automatically by the annotation engine 140 to guide the medical professional on how to either confirm or adjust the lesion annotation. In some instances, the user assistance engine 150 canprovide prompts to the user during any operation associated with a user input 160. In some instances, the prompts can be smart prompts that provide suggestions to, or otherwise informs, the user that the image should adjust the annotations. In some instances, the suggestion can describe how to annotate the images. As will be described, in some embodiments a smart prompt may be generated by the engine 150 based on one or more analyses of the images (e.g., images 130 or 170).

[0156] In this way, the medical professional may prepare the annotated images 170 for the system to use, such as via adjusting the cardiac images, to generate a three-dimensional model 180 and / or index indicative of vascular function. For example, in some embodiments the system may generate a three-dimensional model 180 based, at least in part, on matching features depicted in the cardiac images. The system may determine geometry information for the patient’s heart, such as assigning diameters and / or radii to the identified vessels. Based on the geometry information, the system may determine an FFR value, or FFR values, associated with a lesion. For example, the FFR values may be determined across the length of the vessels such that a drop in FFR across the lesion may be determined.

[0157] In some instances, user assist may provide alerts, such as banners or prompts, during processes by the system 100 besides during the annotation 140 process. For example, user assist may provide alerts regarding techniques to initially mark the lesion within the selected images 130, which can occur before or currently with the annotation 140 process, to generate the automatic vessel annotations. By further example, user assist may provide alerts during the selection of the cardiac images 110 and prior to the annotation 140 regarding techniques to choose the selected images 130 for the coronary assessment. In such instances, the alerts can be smart alerts, such smart banners or smart prompts, that, for example, provide recommendations for and / or against the selection of cardiac images 110. Smart alerts, or alerts, do not necessarily have to be connected to user assist and can be present even if user assist is turned off or disabled. For one example, the system can include smart banners, or banners, that can be provided to the user.User Assistance

[0158] Figure 2 is a block diagram of an example assisted annotation system 200, illustrating the interaction between the annotation engine 140 and user assistance engine 150to generate the annotated images 170. As described above, the annotation engine 140 automatically generates vessel annotations on each of the selected images 130 after the medical professional indicates the location of the lesion in each image by a lesion marker via user input 160. In some instances, the selected images 130 can be an optimal frame selected from a video angiogram. This optimal frame can depict the vessel during an end diastolic phase such that a user can easily select the optimal frame from among the various frames.

[0159] In this assisted annotation system 200, the annotation engine 140 may generate, for example, an ostium annotation 201, lesion annotations 202, selected vessel annotations 203, and / or major side branch annotations 204 on the selected images 130. In some instances, the representations for different types of vessel annotations can be depicted similarly or differently. For example of the annotations (or “vessel markers”), the contours defining the boundaries for the lesion and the selected vessel may both be depicted with red outlines of their perimeters, the vessel ostium can be depicted as a red target, and the major side branches can be depicted with a white outline of their perimeter. The colors of the annotations are not intended to be limiting. For example, the color of the vessel markers for each of the major side branches can differ, be the same, or be the same as the selected vessel.

[0160] In some instances, the annotation can change based on user input, such as changing to a different color or changing the outline to a line along the centerline of a vessel. For example, during the selected vessel annotation 203, when adjusting the length of the selected vessel, the annotation may change to depict a red line along the centerline of the vessel. In such circumstances, the major side branches can also include a centerline, and one or both outlines may not appear while the centerlines are depicted.

[0161] In some instances, end markers can be depicted at each end of a vessel, for example, in some instances, the ends of one or more major side branches may be annotated such as by a yellow dot and / or yellow circle. In such circumstances, the end marker can be used to adjust the length of the major side branch or reconnect the major side branch to a different portion of the selected vessel. In some instances, selecting and / or hovering over the annotation with a user selection indicator can change the type of annotation displayed.

[0162] These annotations are not intended to be limiting and are illustrated as an example; one or all of these annotations may not be present, and / or additional annotations directed to other vascular characteristics may be included. As described above, a medical professionalmay review each of the annotations via the user interface and may optionally adjust the annotations as described above via user input 160. The medical professional may adjust any of the annotations during this process without a required order and can shift back and forth between the selected images 130 displayed on the user interface to adjust their corresponding annotations before completing any particular one of the selected images 130.

[0163] When turning on user assistance, the user assistance engine 150 can generate prompts corresponding to the annotations which are available for the medical professional to adjust. These prompts can include text, images, animation, and / or icons to provide guidance to the medical professional on how to perform or verify that the corresponding annotation is correctly done (e.g., the annotations accurately corresponding to the vessel features represented in the medical images, or at least within a threshold). In this assisted annotation system 200, the user assistance engine 150 can generate an ostium prompt 211 to remind a user to properly annotate the location of the vessel ostium, a lesion prompt 212 to remind a user to properly annotate contours for the boundaries of the portion of the selected vessel near the lesion, a selected vessel prompt 213 to remind a user to properly annotate the contours for the boundaries of the selected vessel, and / or a major side branch prompt 214 to remind a user to mark each major side branch. However, similar to the annotation engine 140, the prompts are not intended to be limiting and are illustrated as an example; one or all of these annotations may not be present and / or additional annotations directed to other vascular characteristics may be included based on the annotations generated by the annotation engine 140. For example, although the current disclosure primarily focuses on the ostium prompt 211, lesion prompt 212, selected vessel prompt 213, and major side branch prompt 214, there can be additional prompts related to any user interaction during the coronary physiology assessment, such as a systolic alert when the user is selecting an optimal frame for a selected image 130. In some instances, the user assistance engine 150 may not generate prompts corresponding to each annotation available for adjustment by the medical professional.

[0164] The user assistance engine 150 can generate the prompts so that they are viewable sequentially. The sequential order of the prompts can be representative of an order or process that a medical professional or user would be trained to operate the system 100 without the user assistance engine 150. For example, the user assistance engine 150 may initially provide the ostium prompt 211 when turned on, then next display the lesion prompt 212, then next displaythe selected vessel prompt 213, then last display the major side branch prompt 214. The prompts can include buttons to either move on to the subsequent prompt or return to the previous prompt. For example, when the lesion prompt 212 is displayed, the medical professional has the option to move on to the selected vessel prompt 213 or return to the ostium prompt 211 via the user input 160. When the last prompt, e.g. the major side branch prompt 214, appears on the user interface, it can include a button to close the prompt. In some instances, the last prompt can include a button to restart back to the first prompt, e.g. the ostium prompt 211. In some instances, any of the prompts can include a button to close the prompts.

[0165] The medical professional can also adjust an annotation corresponding to a displayed prompt, for example to trigger presentation of a subsequent prompt. As an example, when the ostium prompt 211 is displayed, the medical professional performing an adjustment to the ostium annotation 201 can cause the user assistance engine 150 to replace the ostium prompt 211 and display the lesion prompt 212, such as by moving the ostium marker location. Similarly, although a medical professional may annotate any of the annotations even when user assistance is on, user input performing annotation not corresponding to the displayed prompt will not, in some embodiments, trigger the next prompt. As one example, when the ostium prompt 211 is displayed, the medical professional performing an adjustment to the lesion annotations 202, selected vessel annotations 203, or major side branch annotations 204 may not cause any changes to the prompts and the ostium prompt 211 may still be displayed. However, in alternative instances, the medical professional adjusting an annotation not corresponding to the displayed prompt may replace the displayed prompt with the prompt corresponding to the adjusted annotation. For example, if the ostium prompt 211 is displayed and the medical professional adjusts the selected vessel annotations 203, the user assistance engine 150 may replace the displayed ostium prompt 211 and instead display the selected vessel prompt 213. If a medical professional were to then adjust the selected vessel annotations 203 again, the user assistance engine 150 would move on to the major side branch prompt 214, but if the medical professional were to adjust the lesion annotations 202, the user assistance engine 150 would replace the selected vessel prompt 213 with the lesion prompt 212.

[0166] In some instances, rather than starting with the display of the ostium prompt 211 , the first displayed prompt can be based on a memory saved from the last time the system was operated. Accordingly, the last prompt displayed from the memory can be displayed whenaccessing the user assistance engine 150. Alternatively, the first displayed prompt may not appear until a user adjusts an annotation and that first displayed prompt can correspond to that annotation.

[0167] In some instances, and as described in Figures 6-7, the prompts can be supplemented or replaced with smart prompts that indicate technique(s) to annotate the vessels. A smart prompt, as described herein, reflects a suggestion by the system on how it recommends the user annotate the cardiac images 110 using the user interface 101. This suggestion on how to annotate can be a recommendation to adjust the current annotation of the cardiac images 110 because the current annotations may be incorrect without describing how to correct the annotation. In other instances, the recommendation may further describe how to adjust the current annotation to either a correct location, to include all necessary vessels, or to provide correct contours for the boundaries of the vessel. The suggestion can be based on the system analyzing the cardiac images 110, such as by analyzing the geometries of the vessel or comparing the cardiac images 110 to see whether an annotation of a vessel is included in one but not the other. This suggestion can further be based on contextual information associated with the analysis of the cardiac images 110, such as by determining whether any additional annotation may have an effect on the analysis. This effect can include whether the additional annotation could cause a change in determining FFR values of the vessel or would prevent the system from generating a 3-D model. The effect can also be based on a comparison to automatically generated annotation and evaluated to determine if a less accurate result could occur from changing the automatically generated annotation. This accuracy determination can be based on the same criteria used to initially generate the automatic annotation. For example, if the user moves the ostium annotation from the system’s suggested position using the user interface 101, then a smart prompt can be generated to supplement or replace the ostium prompt 211.

[0168] The smart prompts can provide suggestions (or user instructions) to the user on what annotations can help with the analysis - which, when implemented, can increase the accuracy of the measurements associated with a model (such as a 3D model of the vessels) that are generated from the medical images. Furthermore, a lack of smart prompts can indicate to the user that further annotation may not have an impactful effect. For example, if the cardiac image displays a side branch vessel with two branches and there is only a smart promptsuggesting the annotation of one of the branches, then a user can understand that the annotation of that branch may have an impact effect on the analysis of the cardiac images 110, but that further annotation of the other branch may not have an impact effect on the analysis of the cardiac images 110. Thus, the smart prompts can not only reduce the time for the user to decide what to further annotate on the cardiac images 110 but can also reduce time by helping the user determine when further annotations are no longer impactful. In some instances, the system can prevent generation of the model if user input associated with the suggestions have not been received. In some instances, the system can bypass this restriction if it receives user input to disregard the suggestion, but can provide an alert that accuracy may be reduced. In other instances, the system can require that the suggestion be handled by receiving appropriate user input to ensure that the model maintains a high accuracy in its measurements.

[0169] In some instances, the suggestions for the smart prompts can be based on the user assistance engine 150 determining that there are further annotations for the vessels, and more particularly the vessel indicators (or “vessel features”), which can be made. This determination can suggest that the further annotation could have an impactful effect on the analysis of the cardiac images 110. In some instance, the system 100 can identify potential vessel annotations using computer vision techniques (e.g., classical or machine learning techniques, such as neural networks). In such instances, these techniques can be used to determine whether further annotations could have an impactful effect on the analysis of the cardiac images 110.

[0170] As described above, in some instances, a smart prompt for the ostium can be generated based on vessel geometry and / or contextual information for the cardiac images 110, such as when the user adjusts the ostium annotation from an expected position determined by the automatic vessel annotation. In some instances, the conditions to generate the ostium smart prompts can be based on the contrast or other factors related to the selected images 130. For example, a comparison can be made between a portion of the vessel, either to itself or compared with the remainder of the vessel. This comparison could indicate a projected location of the ostium or whether the current ostium annotation 201 is likely to be in a desired location. For instance, a percentage of the vessel (3%, 5%, 10%, 15%, 20%, etc.) starting from where the vessel begins can be analyzed for a width (or diameter) across each portion of the vessel. The number of diameters taken can be associated with the percentage of the vessel analyzed, for example, a given number of diameters determined per width of the vessel, or based on a pre-determined number (e.g., measuring diameters of 10, 50, 100, 250, etc. portions of the sampled vessel evenly or unevenly spaced).

[0171] The above-described diameters or the change of diameters along the sampled vessel can be compared with a threshold vessel diameter to determine if an ostium smart prompt 701 should be displayed. For instance, the threshold can be based on whether there is a quick change between diameters of adjacent positions along the vessel, if there is a slow narrowing or broadening, or a fast narrowing or broadening. Regarding narrowing or broadening, in some instances, the diameters can be compared among the sampled vessel area to determine if there is slow or fast narrowing and subsequent broadening.

[0172] In some instances, the diameters of the sampled portion of the vessel can be compared to diameters of the remaining portion of the vessel, for example if 10% of the beginning of the vessel were sample, a second sampling of diameters could be done of the remaining 90% of the vessel. This second sampling could be at a smaller frequency than the first sampling, for example, 10 sample per percent of the vessel in the first sampling, such as 100 sampled diameters in the first 10% of the vessel, and 1 -2 samples per percent in the second sampling such as 90-180 sampled diameters for the remaining 90% of the vessel. When comparing these two samplings, the rate of narrowing or broadening in the first sampling can be compared with the rate of narrowing or broadening in the second sampling to see the difference of the rate to detect where the ostium is. For example, an expected ostium location can be determined if the rate of narrowing in the first sampling is determined to be slow but faster than the rate of narrowing in the second sampling. In some instances, the threshold can be associated with a metric of central tendency, such as but not limited to an average and / or mean diameters from the first sampling and / or the second sampling. Based on this analysis, a determination can be made on where the expected location of the ostium is and / or if the current ostium annotation 201 is likely to be at the expected location of the ostium to trigger the ostium smart prompt 701, shown in Figure 7A.

[0173] In some instances, directionality can be compensated for the vessel diameters, such as when there are turns of the vessel in the sampled area. In such instances, the sampled diameters may be inaccurate if the diameter is measured using only points that are equally distanced from the beginning of the vessel. In such instances, the points on each side of thevessel used to measure diameter can be adjusted to avoid any large changes in diameters that would not create a large effect on the selection of the next points on each side of the vessel.

[0174] For example, a smart prompt can be generated for a major side branch based on when the system 100 detects the presence of a potential major side branch of the vessel. For instance, the conditions for a major side branch smart prompt 711 (shown in Figure 7B) can be based on but not limited to if the system 100 determines there are missing diagonal arteries (DIAGs), missing marginal branches of the left coronary arteries (MARGs), or missing right coronary artery (RCAs) branches. The system 100 can analyze one of the selected images 130 with network information, such as segmentation masks and regression lines. For example, based on the selected vessel intended to be displayed in image, there can be major side branches that are associated with the selected vessel. Thus, if there is no segmentation masks and / or regression lines associated with the associated major side branch, e.g., no annotation corresponding, then the system 100 can determine that a major side branch is missing to alert the user by generating a major side branch smart prompt 711.

[0175] Additionally or alternatively, the system 100 can analyze all of the selected images 130, such as three angiograms, or a subset of the selected images 130, such as two angiograms, for inconsistencies. For example, as shown in Figure 7B, the automatic vessel annotation can determine where and if a selected vessel and any main branch vessel are depicted in the angiogram. In some instances, if the number of major side branches automatically detected is equal in each angiogram, then no major side branch smart prompt 711 may be depicted, a normal major side branch prompt 214 may be displayed, and / or the major side branch smart prompts 711 may display an acknowledgment or verification that there do not appear to be any missing major side branch vessels. In such instances, the automatic annotation may associate each detected major side branch vessel with a known vessel type, such as first LCX, second LCX, etc., and further compare all selected images 130 to verify if they all depict the same types.

[0176] If the selected images 130 differ in the number of major side branch vessels and / or the types in each, then the system may provide an alert. If the missing vessel in a selected image 130 (and which was automatically annotated in another of the images) is not well- demonstrated, then no major side branch smart prompt 711 may be depicted. Instead, a normal major side branch prompt 214 may be displayed, and / or the major side branch smart prompts711 may display an alert that a vessel may be missing, but that it is not well demonstrated. A vessel may not be considered well-demonstrated if it is not visible in the angiogram, such as when it is covered by another vessel. If the missing vessel is well-demonstrated, then user annotations can be checked to see if any would correspond with the missing vessel. If so, then no major side branch smart prompt 711 may be depicted, and / or a normal major side branch prompt 214 may be displayed. If there are no user annotations that satisfy this, then the major side branch smart prompts 711 can be displayed, for example, displaying that “the LCX may be missing in the third angiogram.” In some instances, the alert can be simpler, such as “a branch vessel may be missing,” or more complex, such as guiding the user on where the missing vessel is in the third angiogram. Similar to other smart prompts, the major side branch smart prompt 711 may continue to be displayed by the user until annotation is added that addresses the cause of the major side branch prompt 711. The cause can be associated with a location on one of the selected images 130 for the missing major side branch. Accordingly, if there is annotation added that does not address the cause, then the major side branch prompt 711 can continue to be displayed.

[0177] Furthermore, the analysis of the selected images 130 can include analyzing if any of the angiograms includes a bifurcation in the vessel near the lesion. By one example, the system 100 can analyze the contrast of the vessel in one of the selected images 130, such as near the lesion, to determine if a major side branch that bifurcates from the selected vessel near the lesion is not annotated. For another example, the automatic annotations (or “projections”) for all the selected images 130 can be analyzed, such that if a bifurcation in the vessel appears near the selected or determined lesion in the projection of one image, then the projections of the remaining selected images 130 can be analyzed to see if they also contain the bifurcation near the lesion. If the vessel is missing in one of the projections of the other images and the missing vessel is well demonstrated, then the major side branch smart prompts 711 can be displayed. Otherwise, a major side branch smart prompt 711 may not be depicted, such that a normal major side branch prompt 214 may be displayed instead. In some instances, the major side branch smart prompts 711 may still be displayed to alert that a vessel may be missing, but it is not well demonstrated, and that verification should be provided that there are no missing major side branch vessels.

[0178] A missing vessel can be considered well-demonstrated when the vessel can be detected by automatic annotations but where the confidence score is not high enough to automatically annotate the vessel. This could happen, for instance, where the vessel overlaps with itself or another vessel, such that the entirety of the vessel cannot be analyzed with a high enough confidence to annotate automatically but where there can still be information extracted about the vessel. For example, information related to the missing vessel can be determined using computer vision based detection and / or deep learning approaches such as regressionbased methods to determine a pathway of the vessel. By another example, segmentation-based methods can be used to determine the shape of the vessel by producing masks. By analyzing the masks, pathways of the vessel and / or shape can be used to generate the alert. These methods may use information in the other selected images 130 to generate the path masks and / or segmentation masks. In some instances, the information gathered can be focused on an area of interest, such as the expected location of the missing vessel based on its relationship to other projected vessels in the other images. For instance, for bifurcations near the vessel, the area of interest can be near the lesion.

[0179] By further example, a lesion smart prompt can be generated to replace the lesion prompt 212 when the system 100 determines that the lesion annotation is recommended based on conditions, which are described in further detail below starting on paragraph

[0163] , For instance, the conditions for a lesion smart prompt to be generated can be based on, but not limited to, if the lesion is not well demonstrated in at least one of the selected images 130, if there are suboptimal contours for the boundaries of the annotations in at least one of the selected images 130, if there are overlap or bifurcations of vessels near the lesion in at least one of the selected images 130, if there is bad contrast in at least one of the selected image 130, or if there is inconsistency on the lesion location compared across the selected images 130. For example, the system 100 can use deep learning approaches and / or computer vision based detection to analyze the boundary contours near the lesion that can be used to trigger a smart lesion prompt. For example, the analysis can be based the contrast analyzed around the gray levels in comparison to the background or vessel-detection filters based on expected characteristics of the vessel (such as shape or curvature). In some instances, the vesseldetection filters can be applied after thresholding the image, such as by assigning each pixel of the image a value as either black (0) or white (1). By another example, the other selectedimages 130 can be used to see if there are differences in the boundary contours of the lesion, such as if one of the selected images 130 has overlaps near the lesion from other vessels or self-overlap, such as from loops. By another example, the analysis can be based on detecting suboptimal contours, such as by comparing the original boundary contours that provided automatically by the system 100 against active contours. These active contours can be based on gradients and / or edge detection determined based on changes made (such as by the user) to the boundary contours from the original annotation. The lesion smart prompt can guide a user on where the expected contours for the boundaries of the lesion are and / or an expected location of the lesion on the selected vessel. Regarding when there is an inconsistency, the system 100 can analyze all of the selected images 130, such as three angiograms, or a subset of the selected images 130, such as two angiograms.

[0180] By further example, a selected vessel smart prompt can be generated to replace the selected vessel prompt 213 when the system 100 determines that the selected vessel annotation is recommended based on conditions, which are described in further detail below starting on paragraph

[0163] , For instance, the conditions for a selected vessel smart prompt can be based on there are portions of the selected vessel that are missing in a first angiogram of the selected images 130 but that appear in a second angiogram of the selected images 130. To generate selected vessel smart prompts, the contour boundaries of the selected vessel can be analyzed similar as described in paragraph

[0156] for the lesion smart prompts. By another example, the system 100 can identify contrast problems by comparing the selected vessels to its background and surroundings, using gradient filters to calculate a contrast level, or by using local thresholding techniques.

[0181] These examples are not intended to be limiting, and in some instances, the conditions described can be applied to the different smart prompts mentioned here or otherwise. For example, smart prompts for marking the lesion prior to the annotation 140 process for automatic annotation can be based on the conditions described to generate the lesion smart prompt.

[0182] The conditions to generate a smart prompt can be based on whether the system determines that further annotation could have an impactful effect on the analysis of the cardiac images. In some instances, the smart prompt may be included to request that evaluate whether the suggested annotation would be beneficial. As such, in some instances, the conditions togenerate a smart prompt can include identifying a portion of the vessel that could be further annotated and determining if some annotation related to that portion might have an impact on the analysis. In some instances, the suggested annotations can be weighed based on the conditions to determine if a smart prompt is generated. For example, a suggestion to annotate a detected branch vessel closer to the lesion may be weighed heavier in favor of generating a smart prompt compared to a suggestion to annotate a detected branch vessel that is further from the lesion, such as near the distal location of the selected vessel. In some instances, these conditions can be a suggestion threshold. In such instances, the system 100 can determine a confidence score or weight of the suggestion to compare against the suggestion threshold.

[0183] In some instances, the system 100 can allow for only manual annotation of the vessels, such as if the user does not mark the lesion for automatic annotation or if automatic annotations are turned off. In such instances, the detection of a portion of the vessel for generating a smart prompt can be based on the same vessel detection process implemented by the annotation engine 140 for automatic vessel detection. For example, if the user does not use the automatic vessel detection and annotating the cardiac images 110 are annotated manually through the user interface 101, then smart prompts can be generated that provide suggestions for vessel annotation that are similar to those that would be automatically annotated. In some instances, there can be smart prompts that suggest vessel annotation that may not be sufficient for automatic annotation. In such instances, the conditions to generate a smart prompt can require less than the conditions to enable automatic annotation. In some instances, the conditions to generate a smart prompt can be the same as the conditions to enable automatic annotation.

[0184] Automatic vessel detection may include conditions that may be met before the system automatically annotates the vessel. For example, the conditions for automation can be based on how confidently the system can identify and detect vessel geometry and / or vessel boundaries, which can be affected by vessel overlap or tortuosity. Even if the system 100 determines that the conditions for automatic annotation are not met, the user assistance engine 150 may still provide a suggestion to the user via a smart prompt. For example, when implementing automatic vessel detection, the system may not automatically detect a side branch already due to overlapping of the side branch with another part of the vessel that causes a lack of confidence in the vessel boundary, but the system may still generate a smart promptsuggesting the annotation of that side branch because it could be impactful in the analysis of the cardiac images 101.

[0185] In some instances, the system 100 can determine a confidence score that can be compared to an annotation threshold to inform whether to automatically annotate. This confidence score can be the same or different than the confidence score compared against the suggestion threshold. In some instances, the suggestion threshold may be lower than the automation threshold, so a vessel may not be automatically annotated but is high enough to provide a smart prompt to the user to add the vessel. For example, a major side branch vessel may have too much overlap and / or tortuosity to confidently detect a marking for and / or contours for its boundaries and therefore may not be able to meet the automation threshold, but it may be enough to meet the suggestion threshold to generate a smart prompt. A user may then determine whether or not to annotate the vessel based on the suggestion to clear the smart prompt. Although marking and / or contour confidences can be used for vessel detection, there may be other confidence scores that can be used for these thresholds and the examples related to the use of markings and / or contours should not be construed to be limiting. By further example, the automation threshold may be related to whether annotation of the suggestion might affect the analysis of the vascular images, such as for determining FFR.

[0186] As described above, a smart prompt can be generated corresponding to each nonsmart prompt, such as prompts 211-214. For example and as shown in Figure 7A, there can be an ostium smart prompt 701 corresponding to the ostium prompt 211. By another example, there can be lesion smart prompt corresponding to the lesion prompt 212. By another example, there can be a selected vessel smart prompt corresponding to the selected vessel prompt 213. By another example and as shown in Figure 7B, there can be major side branch smart prompts 711 corresponding to the major side branch prompt 214. Accordingly, there can be one or more smart prompts for each non-smart prompt, such that the example listed should not be considered exhaustive. In some instances, the first displayed prompt can correspond with one of the smart prompts or the first smart prompt according to the order or process representative of how users would be trained to operate the system 100 described above. In some instances, the next displayed prompt can be the next smart prompt in the aforementioned order or process. In such instances, the next prompt in the order or process, whether a prompt 211-214 or smart prompt, can be displayed in response to selecting the “next button.”

[0187] If a user inputs a corresponding input associated with the smart prompt, such as annotating the vessels according to a threshold associated with the suggestion of the smart prompt or removing the suggestion, then the smart prompt can be removed from the displayed prompts. When replacing a prompt 211-214 with a corresponding smart prompt, if there is more than one smart prompt corresponding with the prompt 211-214, then the prompt 211-214 may only be replaced to display one of the smart prompts. In such instances, the new smart prompt may include an indication that there are other corresponding smart prompts besides the displayed smart prompt. In such instances, when the displayed smart prompt is removed, such as by a corresponding input, the smart prompt can be replaced by either another corresponding smart prompt or the corresponding prompt 211-214.

[0188] In some instances, rather than replacing the smart prompt with either another corresponding smart prompt or corresponding prompt 211-214, the smart prompt can be replaced with the next prompt 211-214 or smart prompt in the order or process. In such instances, the corresponding prompt 211-214 of the previously displayed smart prompt can be available for display, such as by using the “next button” and “previous button.”

[0189] For example, if there are three major side branch smart prompts 711, the major side branch prompt 214 can be replaced by the first of the major side branch smart prompts. In some instances, there may be no indicator that there are two other major side branch smart prompts, and in some instances, there may be an indicator such as a number in the first major side branch smarts prompt that indicates the remaining amount. When the first of the major side branch smart prompts is removed, the second of the major side branch smart prompts can be displayed, and upon its remove, a display of the third of the major side branch smart prompts. Upon the removal of the last of the major side branch smart prompts 711, in some instances, the major side branch prompt 214 can be displayed, in other instances, an ostium smart prompt 701 or ostium prompt 211 can be displayed as the next in the order or process, and in another instance, the next smart prompt in the order or process can be displayed. After the medical professional has finished either verifying or adjusting the annotations of the selected images 130 to generate the annotated images 170, then annotated images 170 can undergo vascular analysis to generate a three-dimensional model 180 or determine an index indicative of vascular function.Flow Diagram

[0190] Figure 3A is a flow diagram for an example process 300 illustrating the interaction between the annotation engine 140 and user assistance engine 150 to generate the annotated images 170. The process 300 may be performed by a system of one or more processors, for example system 100.

[0191] In step 302, a plurality of cardiac images can be accessed. As described above, these cardiac images can be sorted based on the vessels depicted in the cardiac image. A medical professional can select the target vessel and select the cardiac images depicting the target vessel from multiple.

[0192] In step 304, the selected cardiac images can be displayed.

[0193] In step 306, the displayed cardiac images can be annotated, as described above, to illustrate one or more indicators (or “features”) in the depicted vessels in each of the selected cardiac images. These indicators can include the ostium, lesions, target vessels, and major side branches. In some instances, the user may be prompted to indicate the location of the lesion in each of the selected cardiac images, upon which the system can provide automatic annotation of each of the indicators. The user can then verify the annotation or adjust the annotations via the user interface. Alternatively, the annotations can be determined solely by the medical professional.

[0194] In step 308, which can be concurrent to the annotation process in step 306, prompts associated with user assistance can concurrently be displayed on the user interface. There are at least as many indicators for each of the prompts, and the prompts each specifically correspond to a particular indicator. For example, in such instances, if the prompts only include an ostium prompt, a lesion prompt, a target vessel prompt, and a major side branch prompt, then the indicators at least includes the ostium, the lesion, the target vessel, and the major side branches, but can also include branch vessel end markers. The prompts provide training and assistance to the medical professional by guiding the medical professional on how to annotate the cardiac image — whether providing a reminder to verify the annotations or indicating what the annotations represent.

[0195] After the selected cardiac images are done being annotated, the images can be processed and analyzed in step 310 to either generate a 3-D model and / or determine an index indicative of vascular function, such as FFR along the target vessel.

[0196] Figure 3B is a flow diagram for an example process 320 illustrating the interaction between the prompts. The process 320 may be performed by a system of one or more processors, for example system 100. In some instances, the user assistance system will sequentially display prompts in the following order: ostium, lesion, target vessel, and major side branches. This order is not intended to be limiting and there can be more or less prompts.

[0197] In step 322, user assistance can be turned on via the user interface. In some instances, this can trigger step 324 to display the first prompt, e.g. the ostium prompt. Alternatively, the first prompt to display may not be the ostium prompt or it may be based on a memory in the system. If the next user input is a corresponding input, such as selecting the next or adjusting annotations of an indicator associated with the currently displayed prompt, then in step 326, the system may replace the current prompt with the next prompt. For example, replacing the ostium prompt with the lesion prompt.

[0198] If the next user input is a non-corresponding input, e.g., adjusting an annotation of an indicator that is not associated with the currently displayed prompt, then in step 328, the system may replace the current prompt with the prompt associated with indicator whose annotation was just adjusted. In some instances, the system may take no action and may not update the prompt in response to non-corresponding inputs. In such cases, the prompts may only be displayed sequentially forward. However, in some instances, whether allowing noncorresponding inputs or not, there may be a return input that instructs the system to replace the current prompt with the sequentially previous prompt, thus allowing for prompts to also be displayed sequentially backward.

[0199] The medical professional may continue to receive additional prompts either through steps 326 or 328 until they believe they have verified that the annotations are accurate in step 330.Annotation Engine

[0200] Figure 4 A is an example user interface 410 displaying the annotation of the selected images 130 with user assistance turned off. As shown in Figure 4A, the annotation system operates without user assistance prompts when the user assistance indicator 411 is off. The user assistance indicator 411 can be depicted as a message box. In some instances, the user assistance indicator 411 can be depicted with a visual indicator or graphic (e.g., a person iconand / or a light bulb). In some instances, the user assistance indicator 411 is displayed as off when the user assistance indicator 411 is greyed out. The user interface can be displayed on a computing system, as described above, such as by a web browser 400 or other programs. The user interface 410 allows the medical professional to view and adjust the selected images 130A-C in a substantially parallel manner.

[0201] As shown in Figure 4A, three of the selected images 130A-C are displayed on the user interface 410, with each of the selected images 130 including ostium annotation 201, lesion annotations 202, selected vessel annotations 203, and major side branch annotations 204. In the illustrated example, the medical professional has selected the left-most image to provide additional user input too. In response, the left-most image has enlarged to, for example, show detail and provide a visual cue that it can be interacted with.

[0202] In comparison, the center image 130B and right-most image 130C of the selected images 130 may depict the selected vessel from different viewpoints. Information from these viewpoints may be combined to form the three-dimensional model 180. In the upper left, right, and lower left, example medical images are shown. Each medical image may represent an image from a multitude of images which were taken at a particular viewpoint as illustrated at the bottom of each selected image 130.

[0203] In some instances, when the user assistance is off, user assistance prompt may optionally still be displayed below the cardiac image, e.g. as a banner. In some instances, the banner may display the same information as when user assistance is on or may display less information, such as only displaying the text associated with the prompt. In some instances, when user assistance is turned on, as indicated by the training icon, the banner may resize itself into the full prompt. In some instances, the banners can include smart banners corresponding to the smart prompts. In some instances, only smart banners may be displayed when the user assistance is off. In some instances, all smart banners may be available for display even if more than one smart banner corresponds to the same type of prompt. For example, if there were multiple major side branch smart prompts 711, then rather than compiling them into one smart banner for the major side branch that is replaced with the next smart banner when the first is responded to, all of the major side branch smart prompts 711 can be generated as smart banners. By further example, in some instances, if there is a single ostium smart prompt 701 and three major side branch smart prompts 711, there would be a four banners corresponding to theaforementioned prompts with none for the lesion or selected vessel prompts because they did not have suggestions as smart prompts.

[0204] Figure 4B is an example user interface 420 displaying the annotation of the selected images 130 with user assistance turned on. As shown in Figure 4B, the annotation system operates with user assistance prompts when the user assistance indicator 411 is on. In some instances, the user assistance indicator 411 is displayed as on when the user assistance indicator 411 is colored in, as shown in Figure 4B. The medical professional can turn user assistance on and off by selecting the user assistance indicator 411. The medical professional can adjust or define the behavior and / or settings on how the user assist operates either when the system to turned on or off. For example, a new user may keep all available prompts or banners turned when using the system, whereas a more proficient user may turn off non-smart prompts so that only smart prompts can be displayed or turn off any prompts or banners related. The user interface can be displayed on a computing system, as described above, such as by a web browser 400 or other programs. The user interface 420 operates similarly to user interface 410 but includes user-assisted prompts, such as the ostium prompt 211 displayed in Figure 4B, as described above.

[0205] In some instances, the automatic annotation for one of the selected images 130 can occur concurrently with user annotation for a different one of the selected images 130. As shown in Figure 4B, the automatic annotation analysis indicator 422 can illustrate the progress for automatic annotation of the selected image 130B after the lesion has been marked, while at the same time allowing the medical professional to provide user annotations to selected image 130A. The medical professional may even identify a lesion on a third selected image 130C, and as the vessels are being detected on the third selected image 130C, the medical professional may identify a new lesion on the first selected image 130A. Thus, the medical professional may individually interact with and perform on each medical image using the same user interface. Alternatively, the system 100 may require that a user provides the location of the lesion in each of the selected images 130 before performing the automatic annotation.User Assistance Prompts

[0206] Figures 5A-D are close-up example user interfaces displaying various prompts during vessel annotation. Figure 5 A is a close up an example user interface 510 of vesselannotation when the ostium prompt 211 is displayed. Figure 5B is a close up an example user interface 520 of vessel annotation when the lesion prompt 212 is displayed. Figure 5C is a close up an example user interface 530 of vessel annotation when the selected vessel prompt 213 is displayed. Additionally, Figure 5C depicts additional user input options 506 that can allow a medical professional to, for example, change the user input indicator 503 type such as for selection or annotation, delete annotations, change the lesion marker location, and undo changes. Additionally, the additional user input options 506 can include options to zoom in / out, hide annotations, and hide the additional user input options 506. The additional user input options 506 can be available even when user assistance is off. Figure 5D is a close up an example user interface 540 of vessel annotation when the major side branch prompt 214 is displayed. Additionally, the end markers at the end of the major side branch annotations 204 is displayed in Figure 5D.

[0207] Although sequentially ordered, as described above, there are some instances in which the prompts are displayed in a different order, for example, based on adjustments of the annotation by the medical professional. As described above, the user interface can include vessel annotation for the ostium annotation 201, lesion annotations 202, selected vessel annotations 203, and major side branch annotations 204. As described above and shown in Figures 5A-D, ostium annotation 201 can be illustrated as a red target, lesion annotations 202 and selected vessel annotations 203 can be illustrated as red outlines, and major side branch annotations 204 can be illustrated as white outlines. Additionally, lesion annotations 202 can include additional annotations illustrating where the lesion begins and ends (shown as solid green lines) and where the narrowest width of the lesion is (shown as dotted lines), which can be used to determine the lesion percentage drop 504. In some instances, these annotations can also be adjusted by a medical professional, and in some instances, the lesion prompt 212 may include further prompts associated.

[0208] The medical professional can adjust the annotations via controlling a user input indicator 503, such as by a cursor or other means for user selection. The cursor 503 can be used to annotate the vessel, which as described above can cause a change in the displayed prompt, as well as interacting with the buttons on the user interface. For example, as described above, the medical professional can select via the 503 the next button 502 to move on to the next prompt or the back button 505 to return to the previous prompt. In some instances, the numberof prompts can be displayed as the current prompt out of the total number of prompts, e.g. where there are only the ostium, lesion, selected vessel, and major side branch prompts and the currently displayed prompt is the lesion prompt 212, then it can display “2 / 4”. In some instances, the prompts can be represented by icons 702, such as in Figures 7A-B, with a prompt indicator 702 as to which prompt is currently being displayed. For example, the prompts can be represented by unfilled circles, and the currently displayed prompt can be depicted as a filled in circle. When implementing smart prompts, the indicators can be replaced with another indicator to show that it is a smart prompt. For example, a smart prompt can be depicted with the same icon but in a different color, such as a yellow unfilled circle and a yellow filled circle. In some instances, the user selecting one of these icons can replace the currently displayed prompt with the prompt corresponding to the icon selected.

[0209] As shown in Figures 5A-D, the prompts can be displayed on the user interface in a position close to the portion of the vessel annotated corresponding to the prompt. Additionally, as shown in Figures 5A-D, the prompts can be displayed on the user interface based on a determined intelligent location of messages so that the prompts are in a position that does not interfere with the annotation of the vessel corresponding to the prompt. This intelligent location of messages can be determined based on vessel geometry and / or contextual information related to the cardiac images. For example, as shown in Figure 5A, the ostium prompt 211 is close to the ostium annotation 201 but does not cover the ostium annotation 201. Similarly, as shown in Figure 5B, the lesion prompt 212 is positioned near the lesion annotations 202 but minimally covers the outline of the lesion in the lesion annotations 202. In some instances, the prompts can be transparent to increase visibility of the selected images 130 and annotations even when the prompts are displayed. Additionally, the medical professional can select the move button 501 to drag and / or move the prompts to different locations. In some instances, moving the prompt using the move button 501 can lock the prompts into position so that they do not move to new locations when moving changing which prompt is displayed.

[0210] As described above, the prompts can include text and images related to the corresponding annotation, as well as buttons such as move button 501, the next button 502, and user input indicator 503. The text can include instructions such as “Confirm ostial marker is accurately marked,” as shown in ostium prompt 211, “Review lesion boundaries,” as shown in lesion prompt 212, “Ensure RCA is accurately traced throughout its course,” as shown inselected vessel prompt 213 but adjustable for other selected vessels, and “Confirm the major side branches are accurately marked,” as shown in major side branch prompt 214. Additionally, the prompts can include an image or icons depicting the corresponding annotation. For example, the image could be an illustration resembling the annotation, what the annotation looks like, what a correct annotation should look like, or a change to the annotation when adjusting it. In some instances, the images on the prompts can include an animation. This animation could be an example of what adjustments could be made to the annotation or demonstrate how adjustments change the display of the annotation.Smart User Assistance Flow Diagrams

[0211] Figure 6A is a flow diagram of an example process 600 for generating smart prompts. The process 600 may be performed by a system of one or more processors, for example system 100.

[0212] In step 602, the system 100 can analyze the annotations on at least one of the selected images 130 and characteristics identified in the at least one selected images 130. These annotations can preferably be generated by the annotation engine 140, or in some instances, by the user. The annotations can be associated with different indicators for portions of the vessel in the selected images 130. For example, the annotations can be grouped as indicators of the ostium, lesion, selected vessel, and one or more major side branches. As described above, the annotation engine can analyze the selected images 130 to determine how to generate the annotations, such as when the system 100 determines that the probability that an annotation would accurately represent a portion of the vessel. However, in some instances, the analysis of the system 100 may result in a probability that does not provide a high enough confidence to automatically annotate but is high enough to believe user review is necessary. In other instances, this analysis may occur after a user provides an initial manual annotation of the vessels. The system 100 can perform this analysis based on criteria, described in more detail below, associated with each of the indicators, such as by evaluating contrast levels of the selected images 130 or mis-matches between characteristics of the annotations of the same indicator in different of the selected images 130. Accordingly, the analysis can be of a single of or a combination of the selected images 130.

[0213] At step 604, the system 100 can determine suggestions that indicate that the current markers may be incorrect, and in some instances, on how a user can annotate the vessels so that the markers are correctly annotated. These suggestions can be based on the conditions for a smart prompt. In some instances, the analysis by the system 100 to determine automatic annotation and suggestions can occur concurrently.

[0214] The conditions to generate a smart prompt can be based on whether the system determines that further annotation could have an impactful effect on the analysis of the cardiac images. Example conditions to generate a smart prompt can include identifying a portion of the vessel that could be further annotated and determining if some annotation related to that portion might have an impact on the analysis. In some instances, these conditions can be a suggestion threshold. In such instances, the system can determine a confidence score to compare against the suggestion threshold. This confidence score can be the same or different as the confidence score compared with the automation threshold based on vessel geometry and / or boundaries.

[0215] In step 606, the system generates at least one smart prompt that corresponds to at least one of the indicators, which can thus correspond to a non-smart prompt (e.g., prompts 211-214). The smart prompt can be displayed to the user to provide the suggestion. A suggestion, as described throughout the description, can be a suggestion to the user that some characteristic of the indicator needs to be adjusted or corrected. This characteristic can be, for example, that the boundary contours of an indicator (such as the selected vessel or lesion) is incorrectly defined, that an indicator (such as an ostium) is positioned at the wrong location of the selected images 130, and / or that a portion of the indicator (such as the major side branches) is missing an annotation. In other instances, the suggestion may hint how to annotate, such as by providing the type of indicator that has an issue and / or where in the selected image 130 needs to revision or further annotation. In other instances, the suggestion may direct on how the correction to the annotations should be made for a specific indicator, or in other instances, may demonstrate how to make the correction. Similar to the non-specific animation, the suggestion can provide an animation that is specific to the suggestion on how to make the correction, or cause a user element, such as a cursor, to move in association with the suggestion.

[0216] Optionally, prior to step 602, prompts can be first generated, such as ostium prompt 211, lesion prompt 212, selected vessel prompt 213, and major side branch prompt 214, asdescribed above. These prompts can be ordered in a sequence of prompts. Thus, in such instances, the prompts can be replaced by the generated by smart prompts that correspond with the same indicators. Accordingly, an ostium prompt 211 could be replaced by an ostium smart prompt 701, if an ostium smart prompt 701 is generated. In instances where annotation analysis and analysis to determine suggestions occur concurrently, a non-smart prompt may not be displayed initially at all and instead the corresponding smart prompt can be displayed. Accordingly, the non-smart prompts can be sequenced without being displayed, and replaced by the smart prompt in the sequence prior to the display. Thus, the smart prompts can be used to replace the corresponding prompt in the sequence of prompts that are available for display to the user. The sequence of prompts can be displayed to the user after the generated smart prompts replace their corresponding non-smart prompts.

[0217] In step 608, the system 100 can receive a user response associated with the annotation. The user response can be further annotation of the selected images 130. At step 610, the system can determine whether the user input addresses the issue with an annotation associated with the suggestion. Accordingly, the system 100 can rerun the analysis with the new annotation to determine whether there are still issues corresponding with the suggestion. The system 100 can determine if the annotation corresponds with the location of the selected images 130 associated with the suggestion. In some instances, the system 100 can analyze whether the annotations meet a minimum threshold for what the system 100 expects the user annotation to be. During analysis, the system 100 can review only the portion of the selected images 130 that corresponds with the suggestion. For example, if the suggestion is to review the lesion from a smart prompt, then the system 100 may only analyze the portions of the selected images associated with the lesion. By another example, if the suggestion is to check if there is a major side branch missing near the ostium, then the system 100 may only analyze the portions of the selected images near the ostium and / or where there might be a bifurcation from the selected vessel. In other instances, the system 100 may re-analyze all portions of the selected images 130. In such instances, the system 100 may evaluate to determine only whether the currently displayed suggestion is addressed or if all suggestions, including those not currently displayed, are addressed. In some instances, the system 100 may evaluate if the user annotation triggers a new smart prompt.

[0218] If the user response is not associated with the suggestion, then at step 610, the smart prompt 614 can continue to be displayed. Alternatively, a user may remove the suggestion or skips it to another suggestion. A user response can be associated with the suggestion when it addresses the suggestion to at least a minimal degree (e.g., a minimum threshold). In other instances, a user response can be associated with the suggestion when the system receives a annotation approximately in the portion of the selected images 130 associated with the lesion. If the user response is associated with the suggestion, then at step 612, the smart prompt can be removed from display to the user. In some instances, if there is more than one smart prompt corresponding to the same prompt, only the first of the smart prompts may be displayed and the remaining smart prompts may be displayed after clearing the prior smart prompt, such as by annotating the vessel as suggested or removing the smart prompt. For example, if there are two selected vessel smart prompts, then only one of the selected vessel smart prompts may be available for display to the user. After the first selected vessel smart prompt is cleared, then the second smart prompt is displayed. In instances where analysis of the entire selected images 130 occurs for all suggestions, the system 100 may remove the second smart prompt from being displayed if the annotation already is associated with the suggestion of the second smart prompt. In other instances, the system 100 may re-analyze the selected images 130 once the second smart prompt is about to be displayed to determine whether there was already user annotation that was added that is associated with the suggestion of the second smart prompt.

[0219] In instances where non-smart prompts are first sequences or generated, after clearing all smart prompts corresponding to at least one of prompts 211-214, the corresponding prompt 211-214 can be re-added to the sequence of prompts to replace the smart prompt and be available for display to the user. In some instances, the smart prompt may supplement prompts 211-214 and prompts 211-214 may already be sequenced after the smarts prompts so that both may be available for display. In such instances, the smart prompts can be ordered before or after the corresponding prompts 211-214, or all smart prompts can be ordered before or after all of the prompts 211-214. For example, if the user shuffled through the different prompts without adding user annotations, the user can view both the smart prompts and the non-smart prompts. In response to clearing the smart prompt, the smart prompt may be removed rather than replaced with the already available corresponding prompt 211-214.

[0220] Figure 6B is a flow diagram of an example process 620 for generating smart banners when user assistance is turned off. The process 620 may be performed by a system of one or more processors, for example, system 100. The steps provided below can be incorporated into the user-assisted annotation systems described above when the system is off. As described above, when user-assist is turned off, the systems can display banners instead of prompts. The following steps describe smart banners being displayed instead of a banner for each corresponding prompt. In some instances, the steps described below can be used to generate smart prompts when the user assistance is turned on, such as described above, such that only smart prompts are generated without the need to replace non-smart prompts 211-214. In some instances, the smart banners can be utilized even when user assistance is turned on. The smart banners can also be displayed in a similar manner as prompts. Similarly, the prompts, non-smart and smart, can be displayed in a similar manner as banners in some instances.

[0221] In step 622, the system can determine suggestions that a user should annotate vessels, similar to step 604.

[0222] In step 624, a smart banner can be generated based on each of the suggestions and compiled together into a sequence of banners. The ordering of the sequence of banners can be based on the process or order that a physician would be taught to annotate with the system, such as looking at the ostium, lesion, selected vessel, and major side branches. In some instances, the assisted annotation system 620 may include non-smart banners in the sequence of banners and can either all be before or after the smart banners. Alternatively, the smart banners may be ordered before or after the corresponding banners.

[0223] In step 626, the first smart banner ordered first in the sequence of banners can be displayed to the user. In some instances, the first banner in the sequence of banners can be displayed. In some instances, in response to the user turning off smart assist, the banner or smart banner corresponding to the prompt 211-214 or smart prompt displayed prior to turning off smart assist can be the first displayed.

[0224] In step 628, the displayed smart banner can be removed from display in response to a corresponding input to clear the banner. A corresponding input can be associated with the suggestion, such as annotating the vessel in a manner as suggested, selecting an icon by user input 160 to request the display of another banner, such as by the next or previous button, orby removing the smart prompt by user input 160, such as because the user has reviewed and believes annotation as suggested is not necessary. If the corresponding input is associated with the suggestion, then in step 630, the banner can be removed from the sequence of banners.

[0225] After either step 628 or 630, in step 632, another smart banner can be displayed to replace the prior smart banner. In some instances, the prior banner can be replaced with a nonsmart banner. In some instances, if there are no remaining banners in the sequence of banners, then no banner will be displayed. In some instances, the user can select to remove the banners so that the sequence of banners is not displayed.Smart User Assistance Prompts

[0226] Figures 7A-B illustrate detail of example user interfaces displaying various prompts during vessel annotation. Figure 7A is a close up of an example user interface 700 of vessel annotation displaying when the ostium prompt 211 is replaced by an ostium smart prompt 701. Figure 7B is a close up of an example user interface 710 of vessel annotation displaying when the major side branch prompt 214 is replaced by one or more major side branch smart prompts 711.

[0227] Although the smart prompts can be shown in addition to the prompts 211-214, Figures 7A-B illustrate when the corresponding prompts 211-214 are replaced by one or more smart prompts. As shown by the prompt indicator 702, the corresponding prompt can be indicated according to icons to depict which prompt is displayed in the ordering of prompts and how many prompts there are.

[0228] In Figure 7A, non-smart prompts, such as the lesion prompt 212 and selected vessel prompt 213, are indicated as unfilled circles and the smart prompts, such as the ostium smart prompt 701 and major side branch smart prompt 711, can be depicted in a different color (e.g. yellow instead of white). Furthermore, the smart prompts may include addition indicators such as the suggestion indicator 703 or a bar matching the color of the prompt indicator 702 based on whether the prompt is smart or not.

[0229] The smart prompt can provide information such as the current markers do not correctly annotate the vessel, such as “consider adding another side branch.” In some instances, the smart prompt can describe how the user could annotate the system and / or a location where to annotate the vessel, for example, the smart prompt could display “consider adding the sidebranch near the lesion.” The suggestions can include several different categories, such as but not limited to, the corresponding annotation to the vessel and the suggested location for that annotation. The suggested locations can include but are not limited to near the ostium, near the lesion, near the lesion percentage drop 504, near an already annotated branch vessel (which may be further indicated by a color of the annotation), along the branch vessel, along about a certain percent of a certain vessel (e.g., along about 50% of the selected vessel, along about 20% of the green branch vessel starting from the selected vessel, about 75% between the lesion percentage drop 504 and the ostium annotation 201), etc. In some instance, the suggestion of a smart prompt could display the text as the corresponding prompt 211-214 but with an indicator that informs the user that the system believes there may be some issue with the current annotation, rather than just informing the user to review. For example, the smart prompt for the ostium may still display “please review the markings in the ostial area” but include the suggestion indicator 703 or other indicators to differentiate an ostium prompt 211 from an ostium smart prompt 701.Smart User Assistance Banners

[0230] Figures 8A-B illustrate detail of example user interfaces displaying various prompts during vessel annotation. Figure 8A is a close up of an example user interface 800 of vessel annotation displaying smart banners for the ostium when user assistance is turned off. Figure 8B is a close up of an example user interface 810 of vessel annotation displaying smart banners for the major side branch when user assistance is turned off. Figure 8C illustrates detail of an example user interface 820 of vessel annotation displaying smart banners for the selected vessel when user assistance is configured to be off.

[0231] Although smart banners are discussed below, the system can also include banners, such as but not limited to an ostium banner, a lesion banner, a selected vessel banner, and a major side branch banner. The aforementioned banners can be displayed alongside or in lieu of the smart banners described here. Smart banners can be indicated based on a display of the suggestion indicator 703. Figure 8A illustrates an ostium smart banner 801 in which non-smart banners are not depicted. In some instances, banners may be cycled through by selecting on the banners themselves to display the next banner. Alternatively, the banners may include the next button 502 and the back button 505, along with an indicator for the current banner out ofthe total number of banners, as depicted in Figure 8B. In some instances, a lack of such indicators can be because there is only one banner.

[0232] Figure 8C provides an expanded view of a display via the user interface 101, including one of the selected images 130, which displays a selected vessel smart banner 821. It also includes a depiction of a cardiac video that is controllable by the playback options 825 and which includes a determined optimal frame 824 for cardiac analysis. Additionally, another of the selected images 130 displays a lesion marker 823. As described above, an optimal frame can be associated with a prompt or banner (not displayed), regular or smart. The regular systolic alert (e.g. , banner or prompt) can remind the user to verify that they selected an optimal frame. In some instances, a user may select an optimal frame for an angiogram video to be used as the determined optimal frame 824. This optimal frame can depict an end diastolic phase of the displayed vessel. Each frame of the angiogram video can be analyzed, such as by using a neural network, to determine an end diastolic confidence score for each frame. Accordingly, more than one end diastolic frame can exist if more than one than one frame has a confidence score about a confidence threshold. The system can include a visual indicator for the end diastolic frame with the highest end diastolic confidence score, and in some instances, the system may automatically select that frame initially. In further instances, there can be an option to shuffle only between frames whose end diastolic confidence score is above the confidence threshold.

[0233] In other instances, when there is a clear cycle of end-diastolic frames approximately spaced apart, then a smart systolic alert (e.g., banner or prompt) can be generated. This smart systolic alert can inform a user if they selected a frame that has a confidence score below the threshold. Approximately spaced apart frames do not require an exact pattern but could include, for example, 5 non-end-diastolic frames, followed by an end-diastolic frame, then 5 more non-end-diastolic frames, followed by an end-diastolic frame, then 4 more non-end- diastolic frames, etc.. In such instances, if a user-selected frame is not one of the end-diastolic frames, then an alert can be generated that either informs the user to select a different frame or guides the user to select a frame that is end-diastolic. In some instances, the alert can guide the user to the frame with the highest end diastolic confidence score. In some instances, if a user has selected an end-diastolic image, an alert may appear confirming such. Whether a frame is determined to be end diastolic can be based on a comparison of its end diastolic confidencescore to a confidence threshold. This determination can be not end-diastolic, unlikely to be end-diastolic, uncertain if end-diastolic, likely to be end-diastolic, end-diastolic, a combination of, a sub-combination of, etc. The cycle used to generate an alert may consider frames considered to be likely to be end-diastolic as an equivalent to end-diastolic. In instances where there is not a clear cycle then a prompt may not be able to be generated, such as 5 non-end- diastolic frames, an end-diastolic frame, 5 more non-end-diastolic frames, an uncertainty of end-diastolic frame. If a prompt cannot be generated due to a lack of a clear cycle, it may alert the user of such, such as via providing an alert that states, “Systolic alert is not possible for this DICOM due to lack of ED candidates.”

[0234] Furthermore, Figure 8C displays cardiac analysis instructions 822 that can be displayed alongside or separate from the non-smart prompts 211-214 or smart prompts. The cardiac analysis instructions can be statis and remain on display to the user throughout and / or prior to the annotation 140 process.

[0235] The banners can depict the same information or suggestions that was in the corresponding prompts 211-214. The banners can be displayed when the user assistance indicator 411 is toggled off. In some instances, smart banners that correspond to the same prompt may all be displayed as banners in the sequence of banners or there may only be one depicted at a time. For example, Figure 8B depicts a major side branch smart banner 811 that can be followed by another major side branch smart banner 811 if a corresponding input is received by the user interface 101 to show the next banner. The display of the banners can be turned off using the banner exit button 802. In some instances, there can be an additional button to remove the displayed banner from the sequence of banners, such as if the user determines that the suggested annotation is not necessary.Other Embodiments

[0236] All of the processes described herein may be embodied in, and fully automated, via software code modules executed by a computing system that includes one or more computers or processors. The code modules may be stored in any type of non-transitory computer- readable medium or other computer storage device. Some or all the methods may be embodied in specialized computer hardware.

[0237] Many other variations than those described herein will be apparent from this disclosure. For example, depending on the embodiment, certain acts, events, or functions of any of the algorithms described herein can be performed in a different sequence or can be added, merged, or left out altogether (for example, not all described acts or events are necessary for the practice of the algorithms). Moreover, in certain embodiments, acts or events can be performed concurrently, for example, through multi-threaded processing, interrupt processing, or multiple processors or processor cores or on other parallel architectures, rather than sequentially. In addition, different tasks or processes can be performed by different machines and / or computing systems that can function together.

[0238] The various illustrative logical blocks, modules, and engines described in connection with the embodiments disclosed herein can be implemented or performed by a machine, such as a processing unit or processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A processor can be a microprocessor, but in the alternative, the processor can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor can include electrical circuitry configured to process computer-executable instructions. In another embodiment, a processor includes an FPGA or other programmable device that performs logic operations without processing computer-executable instructions. A processor can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor may also include primarily analog components. For example, some or all of the signal processing algorithms described herein may be implemented in analog circuitry or mixed analog and digital circuitry. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a mainframe computer, a digital signal processor, a portable computing device, a device controller, or a computational engine within an appliance, to name a few.

[0239] Conditional language such as, among others, “can,” “could,” “might” or “may,” unless specifically stated otherwise, are understood within the context as used in general toconvey that certain embodiments include, while other embodiments do not include, certain features, elements and / or steps. Thus, such conditional language is not generally intended to imply that features, elements and / or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without user input or prompting, whether these features, elements and / or steps are included or are to be performed in any particular embodiment.

[0240] Disjunctive language such as the phrase “at least one of X, Y, or Z,” unless specifically stated otherwise, is understood with the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (for example, X, Y, and / or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one of Y, or at least one of Z to each be present.

[0241] Any process descriptions, elements or blocks in the flow diagrams described herein and / or depicted in the attached figures should be understood as potentially representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or elements in the process. Alternate implementations are included within the scope of the embodiments described herein in which elements or functions may be deleted, executed out of order from that shown, or discussed, including substantially concurrently or in reverse order, depending on the functionality involved as would be understood by those skilled in the art.

[0242] Unless otherwise explicitly stated, articles such as “a” or “an” should generally be interpreted to include one or more described items. Accordingly, phrases such as “a device configured to” are intended to include one or more recited devices. Such one or more recited devices can also be collectively configured to carry out the stated recitations. For example, “a processor configured to carry out recitations A, B and C” can include a first processor configured to carry out recitation A working in conjunction with a second processor configured to carry out recitations B and C.

[0243] As used herein, the term “about” refers to within ±10%.

[0244] The terms “comprises”, “comprising”, “includes”, “including “, “having “, “such as” and their conjugates mean: “including but not limited to”.

[0245] The words “example” and “exemplary” are used herein to mean “serving as an example, instance or illustration”. Any embodiment described as an “example or “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments and / or to exclude the incorporation of features from other embodiments.

[0246] As used herein the term “method” refers to manners, means, techniques and procedures for accomplishing a given task including, but not limited to, those manners, means, techniques and procedures either known to, or readily developed from known manners, means, techniques and procedures by practitioners of the chemical, pharmacological, biological, biochemical, and medical arts.

[0247] Throughout this application, various embodiments of this invention may be presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the invention. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6, etc., as well as individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range.

[0248] Whenever a numerical range is indicated herein, it is meant to include any cited numeral (fractional or integral) within the indicated range. The phrases “ranging / ranges between” a first indicate number and a second indicate number and “ranging / ranges from” a first indicate number “to” a second indicate number are used herein interchangeably and are meant to include the first and second indicated numbers and all the fractional and integral numerals therebetween.

[0249] It should be emphasized that many variations and modifications may be made to the above-described embodiments, the elements of which are to be understood as being among other acceptable examples. All such modifications and variations are intended to be included herein within the scope of this disclosure.Example Clauses

[0250] Examples of implementations of the present disclosure can be described in view of the following example clauses. The features recited in the below example implementations can be combined with additional features disclosed herein. Furthermore, additional inventive combinations of features are disclosed herein, which are not specifically recited in the below example implementations, and which do not include the same features as the specific implementations below. For sake of brevity, the below example implementations do not identify every inventive aspect of this disclosure. The below example implementations are not intended to identify key features or essential features of any subject matter described herein. Any of the example clauses below, or any features of the example clauses, can be combined with any one or more other example clauses, or features of the example clauses or other features of the present disclosure.

[0251] Clause 1. A method implemented by a system of one or more processors comprising: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising one or more vessels; displaying, via a user interface, at least one medical image of the plurality of medical image data; receiving information annotating, via the user interface, vessel markings for at least one indicator of the one or more vessels of the subject of the displayed at least one medical image; and concurrently displaying, via the user interface, at least one prompt on the at least one medical image by the system, wherein each of the at least one prompt corresponds to user instructions for annotating a specific indicator of the at least one indicators.

[0252] Clause 2. The method of Clause 1, wherein each of the at least one prompt is positioned on the user interface proximate to corresponding indicator.

[0253] Clause 3. The method of Clause 2, wherein the specific indicator is visible when displaying the corresponding prompt on the user interface.

[0254] Clause 4. The method of any one of Clauses 1-3, wherein the at least one prompt is re-positioned responsive to user input via the user interface.

[0255] Clause 5. The method of any one of Clauses 1-4, further comprising selecting a target vessel, wherein the one or more vessels comprises the target vessel and the at least one indicator comprises at least one of an ostium, a lesion, a selected target vessel, or at least one major side branch, and the at least one prompt comprises at least one of an ostium prompt, alesion prompt, a selected vessel prompt, or a major side branch prompt corresponding to each of the at least one indicator.

[0256] Clause 6. The method of Clause 5, wherein the ostium prompt corresponds to at least one of user instructions comprising a suggestion to adjust the annotation of the ostium to a correct location or verification that the annotation of the ostium is in the correct location and wherein the suggestion for the correct location of the ostium is associated with one or more diameters of the target vessel.

[0257] Clause 7. The method of any one of Clause 5 or 6, wherein the major side branch prompt corresponds to user instructions to annotate the at least one major side branch in a first medical image of the at least one medical images, and wherein the user instructions to annotate are associated with a missing major side branch vessel annotated in a second medical image of the at least one medical images and not in the first medical image, wherein the missing major side branch vessel is well- demonstrated in the first medical image.

[0258] Clause 8. The method of any one of Clauses 1-7, wherein receiving information annotating the vessel markings for the at least one indicator of the one or more vessels comprises: receiving user input that annotates a lesion marker on the at least one medical image; generating the vessel markings for each of the at least one indicator by the system in the at least one medical image; and at least one of: verifying accuracy of the generated vessel markings, or receiving user inputs to adjust at least one of the generated vessel markings for each of the at least one indicator.

[0259] Clause 9. The method of Clause 8, wherein displaying, via the user interface, at least one prompt comprises displaying the at least one prompt in a pre-determined order, and wherein the method further comprises displaying a next prompt in response to a corresponding user input received by the user interface, the corresponding user input comprising the user selecting a next button or adjusting the generated annotation of the specific indicator corresponding to the displayed prompt.

[0260] Clause 10. The method of Clause 9, wherein the method further comprises displaying a previous prompt in response to a return user input received by the user interface.

[0261] Clause 11. The method of any one of Clause 9 or 10, wherein the method further comprises, in response to a non- corresponding input: displaying, via the user interface, the prompt corresponding to a non-corresponding indicator, the non-corresponding inputcomprising adjusting, via the user interface, the generated annotation of the non-corresponding indicator, the non-corresponding indicator comprising the at least one indicators that do not correspond to the displayed prompt.

[0262] Clause 12. The method of any one of Clauses 1-11, wherein the user instructions comprises at least one of: text, images, or animation.

[0263] Clause 13. The method of Clause 12, wherein the user instructions comprises an example of an adjusted vessel marking.

[0264] Clause 14. The method of any one of Clauses 1-13, further comprising: generating a three-dimensional model depicting the one or more vessels using the annotated at least one medical image; displaying, via the user interface, the three-dimensional model.

[0265] Clause 15. The method of Clause 14, further comprising: determining an index indicative of vascular function based on the three-dimensional model.

[0266] Clause 16. The method of any one of Clauses 1-15, further comprising receiving, via the user interface, at least one additional prompt by the system for selecting a frame, wherein each of the medical image data in the plurality of medical image data comprises a medical video including a plurality of frames.

[0267] Clause 17. The method of Clause 16, wherein the at least one additional prompt for selecting the frame corresponds to user instructions for selecting an end diastolic frame, wherein the end diastolic frame depicts the one or more vessels during an end diastolic phase, and wherein the at least one additional prompt is received in response to determining the plurality of frames comprise a cycle of systolic and end diastolic frames.

[0268] Clause 18. The method of any one of Clauses 1-17, further comprising receiving, via the user interface, at least one additional prompt by the system for selecting, via the user interface, a target vessel of the vasculature of the subject.

[0269] Clause 19. The method of any one of Clauses 1-18, further comprising receiving, via the user interface, at least one prompt by the system for selecting, via the user interface, a subset of the plurality of medical image data.

[0270] Clause 20. The method of any one of Clauses 1-19, further comprising: receiving user input, via the user interface, to either annotate at least one of the at least one indicators or confirm that the at least one indicators are correctly annotated; and generating a model of atleast one of the vessels based on the displayed at least one medical image and at least one of the information or the user input.

[0271] Clause 21. The method of any one of Clauses 1-20, wherein at least one of the prompt comprises a smart prompt that corresponds to user instructions for a suggestion to annotate an indicator of the at least one indicators.

[0272] Clause 22. The method of Clause 21, further comprising: determining a suggested annotation associated with a correct annotation for the indicator based on one or more analyses of the at least one medical image, wherein the suggestion corresponds with the suggested annotation; receiving user input, via the user interface, to either annotate or verify that the indicator has the correct annotation on the displayed at least one medical image; and generating a model of at least one of the one or more vessel based on the displayed at least one medical image, wherein the indicator having the correct annotation in the at least one medical image is usable to generate the model.

[0273] Clause 23. The method of any one of Clauses 1-22, further comprising: determining a suggestion associated with annotating an indicator of the at least one indicators, and replacing a corresponding prompt associated with the indicator with a smart prompt that corresponds to user instructions based on the suggestion.

[0274] Clause 24. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 1-23.

[0275] Clause 25. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 1-23.

[0276] Clause 26. A method implemented by a system of one or more processors comprising: displaying, via a user interface, at least one medical image of a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising one or more vessels; displaying, via the user interface, over the at least one medical image, vessel markings for at least one indicator of the one or more vessels of the subject; and concurrently displaying, via the user interface, at least one prompt over the at least one medical image by the system, wherein each of the at least one prompt corresponds to instructions for a user to annotate a specific indicator of the at least one indicator.

[0277] Clause 27. The method of Clause 26 wherein displaying the vessel markings for the at least one indicator of the one or more vessels comprises: presenting a lesion marker at a selected position on the at least one medical image in response to receiving user input for the lesion marker; displaying the vessel markings for each of the at least one indicator in the at least one medical image, the vessel markings generated by the system; and displaying a next prompt of the at least one prompts in response to at least one of: receiving user input verifying accuracy of the generated vessel marking for the specific indicator corresponding to a currently displayed prompt, or receiving user input to adjust the generated vessel marking for the specific indicator corresponding to the currently displayed prompt.

[0278] Clause 28. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 26-27.

[0279] Clause 29. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 26-27.

[0280] Clause 30. A method of annotating one or more vessels of a medical image for generating a model of the one or more vessels, comprising: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical image data; receiving information annotating, via the user interface, each of the at least two medical images with at least one marker selected from: (i) an ostial marker corresponding with an ostium of the one or more vessels, (ii) a set of major side branch markers corresponding with a set of major side branches of the one or more vessels, (iii) a lesion marker associated with a lesion of the one or more vessels, the lesion marker associated with a boundary near the lesion, and (iv) a selected vessel marker corresponding with a selected vessel of the one or more vessels, the selected vessel markers associated with a boundary of the selected vessel; displaying, via the user interface, at least one prompt to adjust the at least one marker for a correct annotation or to verify that the at least one marker has the correct annotation; wherein the at least one marker having the correct annotation for each of the at least two medical images is usable to generate the model of the one or more vessels.

[0281] Clause 31. The method of Clause 30, further comprising, in response to receiving user input associated with the prompt to adjust or verify a first marker of the at least one marker for a first medical image of the at least two medical images: removing a first prompt of the at least one prompt from displaying, the first prompt corresponding to the first marker; and displaying, via the user interface, a second prompt corresponding to user instructions to adjust a second marker for a correct annotation in a corresponding second medical image of the at least one medical images or to verify that the second marker has the correct annotation in the second medical image, the second prompt being different from the first prompt, and the second marker being a same type as the first marker.

[0282] Clause 32. The method of any one of Clauses 30-31 , further comprising, in response to receiving user input associated with the prompt to adjust or verify a first marker of the at least one marker for a first medical image of the at least two medical images: removing a first prompt of the at least one prompt from displaying, the first prompt corresponding to the first marker; and displaying, via the user interface, a second prompt corresponding to user instructions to adjust a second marker on the first medical image to a correct annotation or to verify that the second marker is correctly annotated on the first medical image, the second prompt being different from the first prompt, and the second marker being different from the first marker.

[0283] Clause 33. The method of any one of Clauses 30-32, wherein at least one of the at least one prompt is a smart prompt generated based on one or more analyses of at least one medical image of the at least two medical images

[0284] Clause 34. The method of Clause 33, further comprising, in response to receiving user input associated with the prompt to adjust or verify a first marker of the at least one marker for a first medical image of the at least two medical images: removing the smart prompt from displaying, the smart prompt corresponding to the first marker; and displaying, via the user interface, a non-smart prompt corresponding to user instructions to verify that the first marker is correctly annotated on the first medical image, the non-smart prompt being different from the smart prompt and corresponding to the first marker.

[0285] Clause 35. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 30-34.

[0286] Clause 36. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 30-34.

[0287] Clause 37. A method implemented by a system of one or more processors comprising: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising one or more vessels; receiving information annotating one or more indicators for the one or more vessels in at least one medical image of the plurality of medical image data; generating a prompt set comprising a prompt for each of the one or more indicators to verify that each of the one or more indicators has a correct annotation; determining a suggestion associated with the correct annotation of at least one indicator of the one or more indicators based on at least one of the information annotating the one or more indicators and one or more analyses of the at least one medical image; generating a smart prompt based on the suggestion corresponding to user instructions to annotate the at least one indicator for the correct annotation; and replacing a corresponding prompt of the one or more prompt with the smart prompt in the prompt set; wherein the correct annotation of each of the one or more indicators on the at least one medical image is usable to generate a model of the one or more vessels.

[0288] Clause 38. The method of Clause 37, further comprising receiving user input associated with the suggestion, the user input comprising either a user annotation of the at least one indicator associated with the suggestion or a removal of the suggestion.

[0289] Clause 39. The method of Clause 38, further comprising replacing the smart prompt with the corresponding prompt in the prompt set in response to receiving the user input.

[0290] Clause 40. The method of any one of Clauses 37-39, wherein determining a suggestion comprises determining a score associated with the annotation of the at least one of the one or more indicators and comparing the score with a suggestion threshold.

[0291] Clause 41. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 37-40.

[0292] Clause 42. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 37-40.

[0293] Clause 43. A method implemented by a system of one or more processors comprising: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising one or more vessels; receiving information annotating one or more indicators for the one or more vessels in at least one medical image of the plurality of medical image data; determining at least one suggestion associated with a correct annotation of at least one indicator of the one or more indicators based on at least one of the information annotating the one or more indicators and one or more analyses of the at least one medical image; generating a sequence of banners comprising a smart banner for each of the at least one suggestion, each of the smart banners corresponding to user instructions to annotate the at least one indicator for a correct annotation or to verify that the at least one indicator has a correct annotation; displaying a first smart banner of the sequence of banners; and replacing the first smart banner and displaying a second smart banner of the sequence of banners in response to a corresponding input; wherein the correct annotation of each of the one or more indicators on the at least one medical image is usable to generate a model of the one or more vessels.

[0294] Clause 44. The method of Clause 43, wherein the corresponding input comprises one or more of: a user annotation of the one or more vessels associated with a particular suggestion of the at least one suggestion, a user input requesting another banner of the sequence of banners, or a removal of the particular suggestion.

[0295] Clause 45. The method of Clause 44, further comprising removing the first smart banner from the sequence of banners in response to the corresponding input received comprising the user annotation of the one or more vessels associated with the particular suggestion or the removal of the particular suggestion.

[0296] Clause 46. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 43-45.

[0297] Clause 47. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of Clauses 43-45.

[0298] Clause 48. A method of annotating one or more vessels of a medical image for generating a model of the one or more vessels, comprising: accessing a plurality of medicalimage data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical image data each including an ostium of the one or more vessels; receiving information annotating, via the user interface, each of the at least two medical images with an ostial marker corresponding with the ostium; and displaying, via the user interface, an ostium prompt corresponding to user instructions to adjust at least one of the ostial markers to a correct location associated with the ostium in the corresponding medical image of the at least two medical images or to verify that the at least one of the ostial marker is in the correct location; wherein the ostial markers being in the correct location for each of the at least two medical images is usable to generate the model of the one or more vessels.

[0299] Clause 49. The method of Clause 48, wherein the ostium prompt is a smart prompt generated based on one or more analyses of the corresponding medical image of the at least two medical images.

[0300] Clause 50. The method of Clause 49, wherein the smart prompt is based on a comparison of vessel diameters of at least a portion of the one or more vessels to a threshold value associated with vessel diameter.

[0301] Clause 51. The method of Clause 50, wherein the threshold value is based on a sampling frequency of the vessel diameters.

[0302] Clause 52. The method of any one of Clauses 50-51, wherein the threshold value is based on a median of the vessel diameters of a selected vessel of the one or more vessels.

[0303] Clause 53. The method of Clause 52, wherein the portion of the one or more vessels comprises a percent of a length of the selected vessel closest to the ostium, the length being from the ostium to a distal end of the selected vessel.

[0304] Clause 54. The method of Clause 53, wherein the percent is between 5-25%.

[0305] Clause 55. The method of any one of Clauses 52-54, wherein the threshold value is based on the vessel diameters of a remainder of the selected vessel not included in the portion of the one or more vessels.

[0306] Clause 56. The method of Clause 55, wherein the threshold value is based on a metric of central tendency of the remainder of the selected vessel.

[0307] Clause 57. The method of any one of Clauses 50-56, further comprising measuring the vessel diameters of the at least the portion of the one or more vessels.

[0308] Clause 58. The method of Clause 57, wherein a first number of measurements of vessel diameters are measured along a first portion of the one or more vessels, the first number of measurements used in the comparison to the threshold value.

[0309] Clause 59. The method of any one of Clauses 50-58, further comprising determining a directionality associated with the at least the portion of the one or more vessels, and wherein measuring the vessel diameters comprises compensating for the directionality of the one or more vessels.

[0310] Clause 60. The method of any one of Clauses 50-59, wherein the threshold value comprises a threshold rate based on a rate of change between the vessel diameters.

[0311] Clause 61. The method of Clause 60, wherein the threshold rate is associated with at least one of a quick broadening or a quick narrowing.

[0312] Clause 62. The method of any one of Clauses 60-61, wherein the threshold rate is associated with at least one of a slow broadening or a slow narrowing.

[0313] Clause 63. The method of any one of Clauses 60-62, wherein the comparison comprises comparing a rate of change between the vessel diameters of a portion of a selected vessel of the one or more vessels closest to the ostium to the threshold rate.

[0314] Clause 64. The method of any one of Clauses 50-63, wherein the user instructions corresponding to the ostium prompt comprises a suggestion associated with the correct location based on the vessel diameters of at least a portion of the one or more vessels.

[0315] Clause 65. The method of any one of Clauses 48-64, further comprising, in response to receiving the user instructions to adjust or verify a first ostial marker of a first medical image of the at least two medical images: removing a first ostium prompt from displaying; and displaying, via the user interface, a second ostium prompt corresponding to user instructions to adjust a second ostial marker different from the first ostial marker to a correct location associated with the ostium in a corresponding second medical image of the at least one medical images or to verify that the second ostial marker is in the correct location.

[0316] Clause 66. The method of any one of Clauses 48-65, further comprising, in response to receiving the user instructions to adjust or verify the ostial marker for a first medical image of the at least two medical images: removing the ostium prompt from displaying; and displaying, via the user interface, an indicator prompt different from the ostium prompt corresponding to user instructions to adjust an indictor marker different from the ostium markeron the first medical image for a correct annotation or to verify that the indicator marker is correctly annotated.

[0317] Clause 67. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 48-66.

[0318] Clause 68. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 48-66.

[0319] Clause 69. A method of annotating one or more vessels of a medical image for generating a model of the one or more vessels, comprising: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising the one or more vessels; displaying, via a user interface, a first medical image and a second medical image of the plurality of medical image data, the first medical image including a first set of major side branches of the one or more vessels and the second medical image including a second set of major side branches; receiving information annotating, via the user interface, the first medical image with a first set of major side branch markers corresponding with the first set of major side branches and annotating the second medical image with a second set of major side branch markers corresponding with the second set of major side branches; and displaying, via the user interface, a major side branch prompt corresponding to user instructions to adjust the first set of major side branch markers to include requisite markings associated with the first set and second set of major side branches or to verify that the first set of major side branch markers are annotated to include the requisite markings; wherein the first set and the second of major side branch markers being annotated to include the requisite markings in the first medical image and the second medical image is usable to generate the model of the one or more vessels.

[0320] Clause 70. The method of Clause 69, wherein the major side branch prompt is a smart prompt generated based on one or more analyses of at least one of the first medical image and the second medical image.

[0321] Clause 71. The method of Clause 70, wherein the smart prompt is based on a comparison of the first set of major side branch markers with the second set of major side branch markers.

[0322] Clause 72. The method of Clause 71, wherein the requisite markings are associated with a number of major side branches, the comparison is based on a first number of major side branches annotated by the first set of major side branch markers and a second number of major side branches annotated by the second set of major side branch markers, smart prompt being generated if the first number is less than the second number.

[0323] Clause 73. The method of any one of Clauses 71-72, wherein the smart prompt is based on a comparison of a corresponding geometry of each of the major side branches annotated by the first set of major side branch markers and the second set of major side branch markers, wherein a type of major side branch is determined based on the corresponding geometry.

[0324] Clause 74. The method of Clause 73, wherein the comparison between a first set of types of major side branches determined for the first set of major side branch markers and a second set of types determined for the second set of major side branch markers, the smart prompt generated if the first set of types and the second set of types differ.

[0325] Clause 75. The method of any one of Clauses 70-74, wherein the smart prompt corresponds to user instructions to annotate at least one major side branch associated with the second set of major side branch markers and not the first set of major side branch markers.

[0326] Clause 76. The method of Clause 75, wherein the second set of major side branch markers comprises the annotation of a bifurcated major side branch in the second set of major side branches that is not included in the first set of major side branch markers, and the smart prompt corresponds to user instruction to adjust the first set of major side branch markers to include the requisite markings, the requisite markings including the bifurcated major side branch in the first medical image.

[0327] Clause 77. The method of Clause 76, wherein a bifurcation of the bifurcated major side branch is near a lesion depicted in the one or more vessels.

[0328] Clause 78. The method of any one of Clauses 70-77, wherein the information associated with annotating the first medical image comprises a confidence score for each of the major side branches in the first set of major side branches, wherein a major side branch marker is annotated on the major side branch when the associated confidence score is above a confidence threshold.

[0329] Clause 79. The method of Clause 78, wherein the at least one major side branch is well-demonstrated in the first medical image.

[0330] Clause 80. The method of Clause 79, wherein the at least one major side branch is well-demonstrated when the information detects the at least one major side branch but the associated confidence score is below the confidence threshold.

[0331] Clause 81. The method of any one of Clauses 78-80, wherein the confidence score is based on a path associated with each major side branch of the first set of major side branches, the path determined based on a regression model.

[0332] Clause 82. The method of any one of Clauses 78-81, wherein the confidence score is based on a mask associated with each major side branch of the first set of major side branches, the mask determined based on a segmentation model.

[0333] Clause 83. The method of any one of Clauses 69-82, further comprising, in response to receiving the user instructions to adjust or verify the first set of major side branch markers of the first medical image: removing a first major side branch prompt from displaying; and displaying, via the user interface, a second major side branch prompt corresponding to user instructions to adjust the second set of major side branch markers to include the requisite markings or to verify that the second set of major side branch markers are annotated to include the requisite markings.

[0334] Clause 84. The method of any one of Clauses 69-82, further comprising, in response to receiving the user instructions to adjust or verify the first set of major side branch markers for the first medical image: removing the major side branch prompt from displaying; and displaying, via the user interface, an indicator prompt different from the major side branch prompt corresponding to user instructions to adjust an indictor marker different from the first set of major side branch markers on the first medical image for a correct annotation or to verify that the indicator marker is correctly annotated.

[0335] Clause 85. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 69-84.

[0336] Clause 86. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 69-84.

[0337] Clause 87. A method of annotating one or more vessels of a medical image for generating a model of the one or more vessels, comprising: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical image data each including a lesion of the one or more vessels; receiving information annotating, via the user interface, each of the at least two medical images with a lesion marker corresponding with the lesion; and displaying, via the user interface, a lesion prompt corresponding to user instructions to adjust at least one of the lesion markers to include correct contours associated with the lesion in the corresponding medical image of the at least two medical images or to verify that the at least one of the lesion marker has the correct contours; wherein the lesion markers having the correct contours for each of the at least two medical images is usable to generate the model of the one or more vessels.

[0338] Clause 88. The method of Clause 87, wherein the lesion prompt is a smart prompt generated based on one or more analyses of the corresponding medical image of the at least two medical images.

[0339] Clause 89. The method of any one of Clauses 87-88, further comprising, in response to receiving the user instructions to adjust or verify a first lesion marker of a first medical image of the at least two medical images: removing a first lesion prompt from displaying; and displaying, via the user interface, a second lesion prompt corresponding to user instructions to adjust a second lesion marker different from the first lesion marker to include correct contours associated with the lesion in a corresponding second medical image of the at least one medical images or to verify that the second lesion marker has the correct contours.

[0340] Clause 90. The method of any one of Clauses 87-89, further comprising, in response to receiving the user instructions to adjust or verify the lesion marker for a first medical image of the at least two medical images: removing the lesion prompt from displaying; and displaying, via the user interface, an indicator prompt different from the lesion prompt corresponding to user instructions to adjust an indictor marker different from the lesion marker on the first medical image for a correct annotation or to verify that the indicator marker is correctly annotated.

[0341] Clause 91. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 87-90.

[0342] Clause 92. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 87-90.

[0343] Clause 93. A method of annotating one or more vessels of a medical image for a model of the one or more vessels, comprising: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical image data each including a selected vessel of the one or more vessels; receiving information annotating, via the user interface, each of the at least two medical images with a selected vessel marker corresponding with the selected vessel; and displaying, via the user interface, a selected vessel prompt corresponding to user instructions to adjust at least one of the selected vessel marker to include correct contours associated with the selected vessel in the corresponding medical image of the at least two medical images or to verify that the at least one of the selected vessel marker has the correct contours; wherein the selected vessel markers having the correct contours for each of the at least two medical images is usable to generate the model of the one or more vessels.

[0344] Clause 94. The method of Clause 93, wherein the selected vessel prompt is a smart prompt generated based on one or more analyses of the corresponding medical image of the at least two medical images.

[0345] Clause 95. The method of any one of Clauses 93-94, further comprising, in response to receiving the user instructions to adjust or verify a first selected vessel marker of a first medical image of the at least two medical images: removing a first selected vessel prompt from displaying; and displaying, via the user interface, a second selected vessel prompt corresponding to user instructions to adjust a second selected vessel marker different from the first selected vessel marker to include correct contours associated with the selected vessel in a corresponding second medical image of the at least one medical images or to verify that the second selected vessel marker has the correct contours.

[0346] Clause 96. The method of any one of Clauses 93-95, further comprising, in response to receiving the user instructions to adjust or verify the selected vessel marker for a first medical image of the at least two medical images: removing the selected vessel prompt from displaying; and displaying, via the user interface, an indicator prompt different from the selected vessel prompt corresponding to user instructions to adjust an indictor marker different from the selected marker on the first medical image for a correct annotation or to verify that the indicator marker is correctly annotated.

[0347] Clause 97. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 93-96.

[0348] Clause 98. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 93-96.

[0349] Clause 99. A method of annotating one or more vessels of a medical image for generating a model associated with the one or more vessels, the method comprising: accessing a plurality of medical images depicting a portion of a vasculature of a subject, the portion of the vasculature comprising the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical images; obtaining information annotating at least one of the at least two medical images with at least one marker corresponding to one or more vessel features, the at least one marker comprising at least one of: (i) an ostial marker corresponding with an ostium of the one or more vessels, (ii) one or more major side branch markers corresponding with one or more major side branches of the one or more vessels, (iii) a lesion marker corresponding to a lesion of the one or more vessels, the lesion marker associated with a boundary of the one or more vessels near the lesion, and (iv) a selected vessel marker corresponding with a selected vessel of the one or more vessels, the selected vessel markers associated with a boundary of the selected vessel, wherein the one or more vessel features comprises at least one of the ostium, the one or more major side branches, the lesion, and the selected vessel; generating, for display on a user interface, at least one prompt configured to cause adjustment of the at least one marker, the adjustment being configured to correct the annotation or verify the at least one marker; wherein the adjustment is configured to enhance generation of the model.

[0350] Clause 100. The method of Clause 99, further comprising, in response to receiving user input associated with a first prompt of the at least one prompt in a first medical image: removing the first prompt of the at least one prompt from displaying, the first prompt corresponding to a first marker; displaying, via the user interface: a second prompt associated with a second marker in the first medical image, the second marker associated with a different vessel feature from the first marker if user input associated with each of the at least one prompt in the first medical image has not been received; or a third prompt associated with the first marker in a corresponding second medical image if user input associated with each of the at least one prompt in the first medical image has been received.

[0351] Clause 101. The method of any one of Clauses 99-100, wherein at least one of the at least one prompt is a smart prompt generated based on one or more analyses of at least one medical image of the at least two medical images, the smart prompt corresponding to user instructions to adjust the at least one marker for the correct annotation.

[0352] Clause 102. The method of Clause 101, further comprising, in response to not receiving user input corresponding to the user instructions of the smart prompt, preventing generation of the model of the one or more vessels.

[0353] Clause 103. The method of any one of Clauses 101 or 102, wherein the smart prompt is generated based on one or more analyses of the at least one marker annotating at least one of the at least two medical images.

[0354] Clause 104. The method of any one of Clauses 101-103, wherein the one or more analyses include comparing the at least one vessel features between the at least two medical images.

[0355] Clause 105. The method of Clause 104, wherein the one or more vessel features comprises at least the one or more major side branches, wherein the one or more analyses includes comparing at least one of a type or a number of the vessel markers for the one or more major side branches between the at least two medical images, wherein the smart prompt is generated if at least one of the type or the number of the vessel markers is different between the at least two medical images.

[0356] Clause 106. The method of any one of Clauses 101-105, wherein the one or more vessel features comprises at least the ostium, wherein the one or more analyses includesdetermining diameters of the one or more vessels proximal to the ostium, wherein the smart prompt is generated if a change between the diameters is above a threshold.

[0357] Clause 107. The method of any one of Clauses 101-106, wherein the one or more analyses to generate the smart prompt occurs simultaneously with generating the information annotating at least one of the at least two medical images with the at least one marker.

[0358] Clause 108. The method of Clause 107, further comprising determining a confidence score associated with a particular vessel feature of the one or more vessel features based on the at least one medical image, wherein: the one or more analyses occurs in response to determining that the confidence score is below an annotation threshold, or the information associated with annotating the particular vessel feature with the at least one marker is received in response to determining that the confidence score is above the annotation threshold.

[0359] Clause 109. The method of Clause 108, wherein the confidence score is based on contrast of the particular vessel feature determined from the at least one medical image.

[0360] Clause 110. The method of any one of clauses 108-109, wherein the one or more analyses comprises: comparing the confidence score to a suggestion threshold; and in response to the confidence score being above a suggestion threshold, generating the smart prompt.

[0361] Clause 111. The method of any one of Clauses 101-110, further comprising, in response to receiving user input: validating whether the user input is adjusting or verifying a first marker of the at least one marker for a first medical image of the at least two medical images to the correct annotation, the correct annotation associated with the user instructions of the smart prompt; in response to the validation of the user input, removing the smart prompt from displaying, the smart prompt corresponding to the first marker.

[0362] Clause 112. The method of Clause 111, further comprising, in response to removing the smart prompt from displaying, displaying, via the user interface, a non-smart prompt corresponding to user instructions to verify that the first marker is correctly annotated on the first medical image, the non-smart prompt being different from the smart prompt and corresponding to the first marker.

[0363] Clause 113. The method of any one of Clauses 101-112, further comprising, in response to receiving user input associated a first smart prompt corresponding to a first marker for a first medical image: removing the first smart prompt from displaying; and displaying, via the user interface: a second smart prompt corresponding to the first marker for the first medicalimage if user input associated with each of the smart prompt for the first medical image has not been received, the second smart prompt corresponding to different user instructions to adjust the first marker for the correct annotation than the first smart prompt; and a third smart prompt corresponding to a second marker of the at least one marker for the first medical image if a user input associated with each of the smart prompt for the first medical image has been received.

[0364] Clause 114. The method of any one of Clauses 99-113, wherein each of the at least one prompt is positioned on the user interface proximate to corresponding vessel feature.

[0365] Clause 115. The method of any one of Clauses 99-114, wherein the at least one prompts comprise at least one banner.

[0366] Clause 116. The method of any one of Clauses 99-115, wherein the ostial marker corresponds with the ostium of the selected vessels.

[0367] Clause 117. The method of any one of Clauses 99-116, wherein obtaining information annotating the at least two medical images with at least one marker corresponding to one or more vessel features comprises: receiving user input annotating a location of the lesion on at least one of the at least two medical image; and at least one of: automatically generating the at least one marker for the one or more vessel features in the at least one medical image; or receiving user inputs to adjust at least one of the generated vessel markers for each of the at least one feature.

[0368] Clause 118. The method of any one of Clauses 99-117, wherein the at least one prompt comprises at least one of text, images, or animation associated with the at least one marker having the correct annotation.

[0369] Clause 119. The method of any one of Clauses 99-118, wherein the at least one marker having the correct annotation for at least one of the at least two medical images increases an accuracy of a 3D model of the one or more vessels generated from the at least two medical images.

[0370] Clause 120. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Clauses 99-119.-n-

[0371] Clause 121. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Clauses 99-119.

Claims

WHAT IS CLAIMED IS:

1. A method of annotating one or more vessels of a medical image for generating a model associated with the one or more vessels, the method comprising: accessing a plurality of medical images depicting a portion of a vasculature of a subject, the portion of the vasculature comprising the one or more vessels; displaying, via a user interface, at least two medical images of the plurality of medical images; obtaining information annotating at least one of the at least two medical images with at least one marker corresponding to one or more vessel features, the at least one marker comprising at least one of: (i) an ostial marker corresponding with an ostium of the one or more vessels, (ii) one or more major side branch markers corresponding with one or more major side branches of the one or more vessels, (iii) a lesion marker corresponding to a lesion of the one or more vessels, the lesion marker associated with a boundary of the one or more vessels near the lesion, and (iv) a selected vessel marker corresponding with a selected vessel of the one or more vessels, the selected vessel markers associated with a boundary of the selected vessel, wherein the one or more vessel features comprises at least one of the ostium, the one or more major side branches, the lesion, and the selected vessel; generating, for display on a user interface, at least one prompt configured to cause adjustment of the at least one marker, the adjustment being configured to correct the annotation or verify the at least one marker; wherein the adjustment is configured to enhance generation of the model.

2. The method of any one of the preceding claims, further comprising, in response to receiving user input associated with a first prompt of the at least one prompt in a first medical image: removing the first prompt of the at least one prompt from displaying, the first prompt corresponding to a first marker; displaying, via the user interface: a second prompt associated with a second marker in the first medical image, the second marker associated with a different vessel feature from thefirst marker if user input associated with each of the at least one prompt in the first medical image has not been received; or a third prompt associated with the first marker in a corresponding second medical image if user input associated with each of the at least one prompt in the first medical image has been received.

3. The method of any one of the preceding claims, wherein at least one of the at least one prompt is a smart prompt generated based on one or more analyses of at least one medical image of the at least two medical images, the smart prompt corresponding to user instructions to adjust the at least one marker for the correct annotation.

4. The method of Claim 3, further comprising, in response to not receiving user input corresponding to the user instructions of the smart prompt, preventing generation of the model of the one or more vessels.

5. The method of any one of Claims 3 or 4, wherein the smart prompt is generated based on one or more analyses of the at least one marker annotating at least one of the at least two medical images.

6. The method of any one of Claims 3-5, wherein the one or more analyses include comparing the at least one vessel features between the at least two medical images.

7. The method of Claim 6, wherein the one or more vessel features comprises at least the one or more major side branches, wherein the one or more analyses includes comparing at least one of a type or a number of the vessel markers for the one or more major side branches between the at least two medical images, wherein the smart prompt is generated if at least one of the type or the number of the vessel markers is different between the at least two medical images.

8. The method of any one of Claims 3-7, wherein the one or more vessel features comprises at least the ostium, wherein the one or more analyses includes determining diameters of the one or more vessels proximal to the ostium, wherein the smart prompt is generated if a change between the diameters is above a threshold.

9. The method of any one of Claims 3-8, wherein the one or more analyses to generate the smart prompt occurs simultaneously with generating the information annotating at least one of the at least two medical images with the at least one marker.

10. The method of Claim 9, further comprising determining a confidence score associated with a particular vessel feature of the one or more vessel features based on the at least one medical image, wherein: the one or more analyses occurs in response to determining that the confidence score is below an annotation threshold, or the information associated with annotating the particular vessel feature with the at least one marker is received in response to determining that the confidence score is above the annotation threshold.

11. The method of Claim 10, wherein the confidence score is based on contrast of the particular vessel feature determined from the at least one medical image.

12. The method of any one of claims 10-11, wherein the one or more analyses comprises: comparing the confidence score to a suggestion threshold; and in response to the confidence score being above a suggestion threshold, generating the smart prompt.

13. The method of any one of Claims 3-12, further comprising, in response to receiving user input: validating whether the user input is adjusting or verifying a first marker of the at least one marker for a first medical image of the at least two medical images to the correct annotation, the correct annotation associated with the user instructions of the smart prompt; in response to the validation of the user input, removing the smart prompt from displaying, the smart prompt corresponding to the first marker.

14. The method of Claim 13, further comprising, in response to removing the smart prompt from displaying, displaying, via the user interface, a non-smart prompt corresponding to user instructions to verify that the first marker is correctly annotated on the first medical image, the non-smart prompt being different from the smart prompt and corresponding to the first marker.

15. The method of any one of Claims 3-14, further comprising, in response to receiving user input associated a first smart prompt corresponding to a first marker for a first medical image:removing the first smart prompt from displaying; and displaying, via the user interface: a second smart prompt corresponding to the first marker for the first medical image if user input associated with each of the smart prompt for the first medical image has not been received, the second smart prompt corresponding to different user instructions to adjust the first marker for the correct annotation than the first smart prompt; and a third smart prompt corresponding to a second marker of the at least one marker for the first medical image if a user input associated with each of the smart prompt for the first medical image has been received.

16. The method of any one of the preceding claims, wherein each of the at least one prompt is positioned on the user interface proximate to corresponding vessel feature.

17. The method of any one of the preceding claims, wherein the at least one prompts comprise at least one banner.

18. The method of any one of the preceding claims, wherein the ostial marker corresponds with the ostium of the selected vessels.

19. The method of any one of the preceding claims, wherein obtaining information annotating the at least two medical images with at least one marker corresponding to one or more vessel features comprises: receiving user input annotating a location of the lesion on at least one of the at least two medical image; and at least one of: automatically generating the at least one marker for the one or more vessel features in the at least one medical image; or receiving user inputs to adjust at least one of the generated vessel markers for each of the at least one feature.

20. The method of any one of the preceding claims, wherein the at least one prompt comprises at least one of text, images, or animation associated with the at least one marker having the correct annotation.

21. The method of any one of the preceding claims, wherein the at least one marker having the correct annotation for at least one of the at least two medical images increases anaccuracy of a 3D model of the one or more vessels generated from the at least two medical images.

22. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of the preceding claims.

23. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of the preceding claims.

24. A method implemented by a system of one or more processors comprising: accessing a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising one or more vessels; displaying, via a user interface, at least one medical image of the plurality of medical image data; receiving information annotating, via the user interface, vessel markings for at least one indicator of the one or more vessels of the subject of the displayed at least one medical image; and concurrently displaying, via the user interface, at least one prompt on the at least one medical image by the system, wherein each of the at least one prompt corresponds to user instructions for annotating a specific indicator of the at least one indicators.

25. The method of Claim 24, wherein each of the at least one prompt is positioned on the user interface proximate to corresponding indicator.

26. The method of any one of Claims 24 or 25, wherein the specific indicator is visible when displaying the corresponding prompt on the user interface.

27. The method of any one of Claims 24-26, wherein the at least one prompt is repositioned responsive to user input via the user interface.

28. The method of any one of Claims 24-26, further comprising selecting a target vessel, wherein the one or more vessels comprises the target vessel and the at least one indicator comprises at least one of an ostium, a lesion, a selected target vessel, or at least one major side branch, and the at least one prompt comprises at least one of an ostium prompt, alesion prompt, a selected vessel prompt, or a major side branch prompt corresponding to each of the at least one indicator.

29. The method of Claim 28, wherein the ostium prompt corresponds to at least one of user instructions comprising a suggestion to adjust the annotation of the ostium to a correct location or verification that the annotation of the ostium is in the correct location and wherein the suggestion for the correct location of the ostium is associated with one or more diameters of the target vessel.

30. The method of any one of Claims 28 or 29, wherein the major side branch prompt corresponds to user instructions to annotate the at least one major side branch in a first medical image of the at least one medical images, and wherein the user instructions to annotate are associated with a missing major side branch vessel annotated in a second medical image of the at least one medical images and not in the first medical image, wherein the missing major side branch vessel is well- demonstrated in the first medical image.

31. The method of any one of Claims 24-30, wherein receiving information annotating the vessel markings for the at least one indicator of the one or more vessels comprises: receiving user input that annotates a lesion marker on the at least one medical image; generating the vessel markings for each of the at least one indicator by the system in the at least one medical image; and at least one of: verifying accuracy of the generated vessel markings, or receiving user inputs to adjust at least one of the generated vessel markings for each of the at least one indicator.

32. The method of Claim 31 , wherein displaying, via the user interface, at least one prompt comprises displaying the at least one prompt in a pre-determined order, and wherein the method further comprises displaying a next prompt in response to a corresponding user input received by the user interface, the corresponding user input comprising the user selecting a next button or adjusting the generated annotation of the specific indicator corresponding to the displayed prompt.

33. The method of Claim 32, wherein the method further comprises displaying a previous prompt in response to a return user input received by the user interface.

34. The method of any one of Claims 32 or 33, wherein the method further comprises, in response to a non-corresponding input: displaying, via the user interface, the prompt corresponding to a noncorresponding indicator, the non- corresponding input comprising adjusting, via the user interface, the generated annotation of the non-corresponding indicator, the non-corresponding indicator comprising the at least one indicators that do not correspond to the displayed prompt.

35. The method of any one of Claims 24-34, wherein the user instructions comprises at least one of: text, images, or animation.

36. The method of Claim 35, wherein the user instructions comprises an example of an adjusted vessel marking.

37. The method of any one of Claims 24-36, further comprising: generating a three-dimensional model depicting the one or more vessels using the annotated at least one medical image; displaying, via the user interface, the three-dimensional model.

38. The method of Claim 14, further comprising: determining an index indicative of vascular function based on the three- dimensional model.

39. The method of any one of Claims 24-38, further comprising receiving, via the user interface, at least one additional prompt by the system for selecting a frame, wherein each of the medical image data in the plurality of medical image data comprises a medical video including a plurality of frames.

40. The method of Claim 39, wherein the at least one additional prompt for selecting the frame corresponds to user instructions for selecting an end diastolic frame, wherein the end diastolic frame depicts the one or more vessels during an end diastolic phase, and wherein the at least one additional prompt is received in response to determining the plurality of frames comprise a cycle of systolic and end diastolic frames.

41. The method of any one of Claims 24-40, further comprising receiving, via the user interface, at least one additional prompt by the system for selecting, via the user interface, a target vessel of the vasculature of the subject.

42. The method of any one of Claims 24-41, further comprising receiving, via the user interface, at least one prompt by the system for selecting, via the user interface, a subset of the plurality of medical image data.

43. The method of any one of Claims 24-42, further comprising: receiving user input, via the user interface, to either annotate at least one of the at least one indicators or confirm that the at least one indicators are correctly annotated; and generating a model of at least one of the vessels based on the displayed at least one medical image and at least one of the information or the user input.

44. The method of any one of Claims 24-43, wherein at least one of the prompt comprises a smart prompt that corresponds to user instructions for a suggestion to annotate an indicator of the at least one indicators.

45. The method of Claim 44, further comprising: determining a suggested annotation associated with a correct annotation for the indicator based on one or more analyses of the at least one medical image, wherein the suggestion corresponds with the suggested annotation; receiving user input, via the user interface, to either annotate or verify that the indicator has the correct annotation on the displayed at least one medical image; and generating a model of at least one of the one or more vessel based on the displayed at least one medical image, wherein the indicator having the correct annotation in the at least one medical image is usable to generate the model.

46. The method of any one of Claims 24-45, further comprising: determining a suggestion associated with annotating an indicator of the at least one indicators, and replacing a corresponding prompt associated with the indicator with a smart prompt that corresponds to user instructions based on the suggestion.

47. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Claims 24-46.

48. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Claims 24-46.

49. A method implemented by a system of one or more processors comprising: displaying, via a user interface, at least one medical image of a plurality of medical image data depicting a portion of a vasculature of a subject, the portion of the vasculature comprising one or more vessels; displaying, via the user interface, over the at least one medical image, vessel markings for at least one indicator of the one or more vessels of the subject; and concurrently displaying, via the user interface, at least one prompt over the at least one medical image by the system, wherein each of the at least one prompt corresponds to instructions for a user to annotate a specific indicator of the at least one indicator.

50. The method of Claim 49 wherein displaying the vessel markings for the at least one indicator of the one or more vessels comprises: presenting a lesion marker at a selected position on the at least one medical image in response to receiving user input for the lesion marker; displaying the vessel markings for each of the at least one indicator in the at least one medical image, the vessel markings generated by the system; and displaying a next prompt of the at least one prompts in response to at least one of: receiving user input verifying accuracy of the generated vessel marking for the specific indicator corresponding to a currently displayed prompt, or receiving user input to adjust the generated vessel marking for the specific indicator corresponding to the currently displayed prompt.

51. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the one or more processors to perform the method of any one of Claims 49-50.

52. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the system to perform the method of any one of Claims 49-50.

Images