The present disclosure is directed to a method and device for automatically predicting FFR based on images of vessel. The method for automatically predicting FFR based on images of a vessel. The method comprises a step of receiving the images of a vessel acquired by an imaging device. Then, a sequence of flow speeds at a sequence of positions on a centerline of the vessel is acquired by a processor. A sequence of first features at the sequence of positions on a centerline of the vessel are acquired by the processor, by fusing structure-related features and flow speeds and using a convolutional neural network. Then, a sequence of FFR at the sequence of positions is determined by the processor through using a sequence-to-sequence neural network on the basis of the sequence of first features.

Disclosed herein are methods and systems for capillary oximetry (e.g., retinal capillary oximetry) using optical coherence tomography (OCT). The method may include obtaining an OCT angiography dataset, performing capillary segmentation based on the OCT angiography dataset to obtain capillary segments, resampling, registering, and/or averaging B-scans of the OCT angiography dataset that correspond to a first capillary segment of the capillary segments to obtain an averaged B-scan for the first capillary segment, determining an anterior and posterior border of the first capillary segment, and determining an oxygen saturation of the first capillary segment based on the averaged B-scan, the anterior border, and the posterior border. Other embodiments may be described and claimed.

Described here are systems and methods for generating and analyzing co-expression signature data from scalar or multi-dimensional data fields contained in or otherwise derived from imaging data acquired with a medical imaging system. A similarity metric, such as an angular similarity metric, is computed between the data field components contained in pairs of voxels in the data field data. The data fields can be scalar fields, vector fields, tensor fields, or other higher-dimensional data fields. A probability distribution of these similarity metrics can be generated and used as co-expression signature data that indicate pairwise disparities in the data field data.

Systems and methods for estimating arterial flow information can include a processor generating a time attenuation sequence for each point of a pair of points along a segment of a coronary artery structure. The processor can determine the arterial flow velocity between the pair of points using the distance between the pair of points and the difference between average transit times associated with the pair of points. The one or more processors can determine the average transit times across the same time window. The processor can determine the arterial flow velocity between the pair of points using the distance between the pair of points and the difference between a first time duration that a number of particles take to pass by a first point of the pair of points and a second time duration that the number of particles take to pass by the other point.

A system and method for measuring health parameters of a subject is disclosed. The system and method are based on a mirror; an image acquisition unit configured with the mirror, and comprising a thermal sensor for capturing thermal images or videos of a body part of the subject; and a processing unit to receive data packets associated with the captured thermal images or videos from the image acquisition unit to identify a region of interest of the body part in each frame of the captured thermal images and videos. Further, the processing unit extracts attributes associated with a heat intensity variation from the identified region of interest region, and compares the extracted attributes with a predetermined set of reference data to measure risk scores associated with the health parameters of the subject based on the comparison. The measured risk scores are displayed by a display unit.

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.

An automated workflow receives a patient study comprising cardiac biomarker measurements and a plurality of echocardiographic images taken by an ultrasound device of a patient heart. A filter separates the plurality of echocardiogram images by 2D images and Doppler modality images based on analyzing image metadata. The 2D images are classified by view type, and the Doppler modality images are classified by view type. The cardiac chambers are segmented in the 2D images, and the Doppler modality images are segmented to generate waveform traces, producing segmented 2D images and segmented Doppler modality images. Using both the sets of images, measurements of cardiac features for both left and right sides of the heart are calculated. The cardiac biomarker measurements and the calculated measurements are compared with international cardiac guidelines to generate conclusions, and a report is output showing the measurements that fall within or outside of the guidelines.

Provided are a diagnostic device and a diagnostic method for predicting fractional flow reserve (FFR) and diagnosing coronary artery lesions via a coronary angiography-based machine learning algorithm. A deep learning-based diagnostic method for diagnosing an ischemic lesion includes: obtaining an angiography image of a patient's blood vessel; extracting a region of interest (ROI) from the angiography image; acquiring diameter information of the blood vessel in the ROI; extracting morphological features of the blood vessel based on the diameter information; and obtaining a predictive FFR value by inputting the morphological features to an artificial intelligence (AI) model and determining whether a lesion is an ischemic lesion.

A method for analyzing a surgical textile. A depth image and a color image of the surgical textile is acquired. One or more processors analyze a depth map of the depth image to apply one or more image classifiers. Blood information is extracted from pixels of the color image if the depth image satisfies the classifier(s). The classifier(s) may be a perimeter classifier, a planarity classifier, a normality classifier, a distance classifier, and/or a color classifier, among others. The distance classifier may include transforming pixel dimensions of a selected surface in the depth image into real dimensions based on distance values of the pixels. A graphical representation may be displayed to indicate the need to move the surgical textile closer or farther to satisfy the distance classifier. A prompt may be displayed to manipulate the surgical textile prior to the steps of acquiring the depth image and the color image.

A method for performing flow analysis in a target volume of a moving organ having a long axis, such as the heart, from 4D MR Flow volumetric image data set of such organ, wherein such data set comprises structural information and three-directional velocity information of the target volume over time, the devices, program products and methods comprising, under control of one or more computer systems configured with specific executable instructions: a) deriving from the 4D MR Flow volumetric image data set at least one derived image data set related to the long axis of the moving organ, for example, by using a multi planar reconstruction; b) determining at least one feature of interest in the 4D MR Flow volumetric image data set or in said derived image data set. The feature of interest may be determined, for example, by receiving input from a user or by performing automatic detection steps on the 4D MR Flow volumetric image data set; c) tracking the feature of interest within the 4D MR Flow volumetric image data set or in the derived image data set; d) determining the spatial orientation over time of a plane containing the feature of interest in the 4D MR Flow volumetric image data set; e) performing quantitative flow analysis using velocity information on the plane as determined in step d). A corresponding device and computer program are also disclosed.