Embodiments of the present disclosure include a method, device and computer readable medium involving receiving image data of one or more coronary arteries, generating a binary segmentation indicating presence of calcium in the one or more coronary arteries from the image data, generating a branch density of the one or more coronary arteries, and assigning a coronary artery label from the branch density to the binary segmentation such that at least one indication of presence of calcium of the binary segmentation is labeled as present in a specific one of the one or more coronary arteries.

Embodiments of the present disclosure relate to methods, apparatus, and computer readable medium for detecting an immune factor responsive to a contrast agent. In some embodiments, pre-contrast image data, early timepoint image data, and delayed timepoint image data are received. In some embodiments, an early timepoint map is generated using a comparison of the pre-contrast image data to the early timepoint image data, and a delayed timepoint map is generated using a comparison of the pre-contrast image data to the delayed timepoint image data, where the early timepoint map represents intravascular contrast, the delayed timepoint map represents the intravascular contrast and extravascular contrast, and the extravascular contrast indicates the immune factor. In some embodiments, a combined immune factor map is generated by voxel-wise subtraction of the early timepoint map from the delayed timepoint map.

Systems and methods for retraining a trained machine learning model are provided. One or more input medical images are received. Measures of interest for a primary task and a secondary task are predicted from the one or more input medical images using a trained machine learning model. The predicted measures of interest for the primary task and the secondary task are output. User feedback on the predicted measure of interest for the secondary task is received. The trained machine learning model is retrained for predicting the measures of interest for the primary task and the secondary task based on the user feedback on the output for the secondary task.

A method for determining a predicted risk level of a clinical endpoint for a predetermined time period for a patient is provided by the present disclosure. The method includes receiving video frames of a heart, the video frames being associated with the patient, receiving electronic health record data including a number of variables associated with the patient, providing the video frames and the electronic health record data to the trained neural network, receiving a risk score from the trained neural network, and outputting a report based on the risk score to at least one of a display or a memory.

A method for determining a predicted risk level of a clinical endpoint for a predetermined time period for a patient is provided by the present disclosure. The method includes receiving video frames of a heart, the video frames being associated with the patient, receiving electronic health record data including a number of variables associated with the patient, providing the video frames and the electronic health record data to the trained neural network, receiving a risk score from the trained neural network, and outputting a report based on the risk score to at least one of a display or a memory.

Disclosed herein are methods and systems for optical coherence tomography (OCT) angiography (OCTA). An interleaved scanning pattern is described herein for both raster and bidirectional scanning methods. The interleaved scanning pattern provides B-scans with different scanning intervals. OCTA images based on the B-scans may be combined to obtain a high dynamic range (HDR) OCTA image. Other embodiments may be described and claimed.

[Problem] The present invention provides analysis technology relating to video data of a fluorescent contrast agent shot by a microscope during an operation, and addresses the problem of providing a method and system allowing information such as BV, BF and MTT, and vascular wall thickness, to be estimated by fluorescent contrast agent analysis, by applying perfusion analysis methods, which allow estimation of information such as BV, BF and MTT, to fluorescent contrast agent analysis.

[Solution] The method for image processing of intravascular hemodynamics according to the present invention is characterized by shooting video using infrared light, wherein the object of shooting is a portion of a blood vessel injected with a standard amount of a fluorescent contrast agent; performing image analysis of a shape of a chronological change curve of intensity values which are image outputs from the video shooting; and calculating relative data for blood volume and blood flow based on results of the image analysis.

Systems and methods are disclosed for assessing organ and/or tissue transplantation by estimating blood flow through a virtual transplant model by receiving a patient-specific anatomical model of the intended transplant recipient; receiving a patient-specific anatomical model of the intended transplant donor, the model including the vasculature of the organ or tissue that is intended to be transplanted to the recipient; constructing a unified model of the connected system post transplantation, the connected system including the transplanted organ or tissue from the intended transplant donor and the vascular system of the intended transplant recipient; receiving one or more blood flow characteristics of the connected system; assessing the suitability for an actual organ or tissue transplantation using the received blood flow characteristics; and outputting the assessment into an electronic storage medium or display.

Optical Coherence Tomography Angiography (OCTA) image representation is obtained having OCTA pixels assigned respective OCTA values. A vessel density map is computed from the OCTA image representation. A fractional deviation map and/or a pattern deviation map is computed for the patient from the vessel density map and a normative database, wherein: (1) the fractional deviation map represents a percent loss of vessel density at each pixel location relative to an expected value based on the normative database; and (2) computing the pattern deviation map includes: computing a pattern map of the vessel density representing a normalized vessel density pattern of the vessel density map relative to an average value of the vessel density map; and computing the pattern deviation map using the pattern map. A loss is determined by using at least one of the fractional deviation map and the pattern deviation map. Other features are also provided.

In hemodynamic determination in medical imaging, the classifier is trained from synthetic data rather than relying on training data from other patients. A computer model (in silico) may be perturbed in many different ways to generate many different examples. The flow is calculated for each resulting example. A bench model (in vitro) may similarly be altered in many different ways. The flow is measured for each resulting example. The machine-learnt classifier uses features from medical scan data for a particular patient to estimate the blood flow based on mapping of features to flow learned from the synthetic data. Perturbations or alterations may account for therapy so that the machine-trained classifier may estimate the results of therapeutically altering a patient-specific input feature. Uncertainty may be handled by training the classifier to predict a distribution of possibilities given uncertain input distribution. Combinations of one or more of uncertainty, use of synthetic training data, and therapy prediction may be provided.