Described herein are technologies for facilitating reconstruction of flow data. In accordance with one aspect, the framework receives a four-dimensional projection image dataset and registers one or more pairs of temporally adjacent projection images in the image dataset. Two-dimensional flow maps may be determined based on the registered pairs. The framework may then sort the two-dimensional flow maps according to heart phases, and reconstruct a three-dimensional flow map based on the sorted two-dimensional flow maps.

There is provided a method of generating a dataset of synthetic images, comprising: for each real image each depicting a real human anatomical structure: extracting and preserving a real anatomical structure region(s) from the real image, generating a synthetic image comprising a synthetic human anatomical structure region and the preserved real anatomical structure region(s), designating pairs of images, each including the real image and the synthetic image, feeding the pair into a machine learning model trained to recognize anatomical structure parts to obtain an outcome of a similarity value denoting an amount of similarity between the real image and the synthetic image, verifying that the synthetic image does not depict the real human anatomical structure when the similarity value is below a threshold, wherein an identity of the real human anatomical structure is non-determinable from the synthetic image, and including the verified synthetic image in the dataset.

An apparatus for assessing a patient's vasculature and a corresponding method are provided, in which the bifurcations in a vessel of interest are identified on the basis of a local change in at least one geometric parameter value of the vessel of interest and the fluid dynamics inside the vessel of interest are adjusted to take account for said bifurcations.

A computerized system and method of modeling myocardial tissue perfusion can include acquiring a plurality of original frames of magnetic resonance imaging (MRI) data representing images of a heart of a subject and developing a manually segmented set of ground truth frames from the original frames. Applying training augmentation techniques to a training set of the originals frame of MRI data can prepare the data for training at least one convolutional neural network (CNN). The CNN can segment the training set of frames according to the ground truth frames. Applying the respective input test frames to a trained CNN can allow for segmenting an endocardium layer and an epicardium layer within the respective images of the input test frames. The segmented images can be used in calculating myocardial blood flow into the myocardium from segmented images of the input test frames.

A non-invasive method of assessing a coronary stenosis or other blockage in an artery or other vasculature is based on determination of a blood flow characteristic. In some embodiments, the blood flow characteristic is a fractional flow reserve determined using a statistical correlation of experimentally determined physiological factors and anatomical factors. In other embodiments, the blood flow characteristic is a blood flow rate determined using machine learning techniques. In further embodiments, the blood flow rate determined using machine learning techniques is a physiological factor used in determining the fractional flow reserve.

The system comprises an image acquisition device for collecting medical images; an image pre-processing device for enhancing the visual quality; an image segmentation device for extracting the region of interest from the image’s background by identifying each image’s pixel characteristics, and dividing the image into segments; a feature extraction and selection device for extracting a set of features and selecting the optimized features; a model training device for training a fuzzy logic-based prediction model and a plurality of diagnosis-specific treatment response models to predict treatment response; and a central processing device coupled to a user input device for receiving a subject patient dataset including features obtained for a reduced feature dataset and comparing the subject patient dataset to a feature data scheme for predicting a response for the subject patient thereby predicting the diseases in its early stage.

A method for analyzing a patient based on a volumetric pulmonary scan includes receiving volumetric pulmonary scan data representative of a patient's pulmonary structure. This method also includes determining a level of transpulmonary pressure defining an effort metric based on one or more characteristics from the received volumetric pulmonary scan data. This method further includes determining one or more physiological or anatomical parameters associated with the transpulmonary pressure based on the received volumetric pulmonary scan data and the effort metric. A non-transitory computer readable medium can be programmed with instructions for causing one or more processors to perform the method for analyzing a patient based on a volumetric pulmonary scan.

The present application provides a method and an apparatus for accurately extracting a vascular centerline, analysis system, and storage medium. The method for accurately extracting vessel centerline comprises: selecting a frame of a two-dimensional coronary angiography image when a blood vessel is fully filled with a contrast agent; acquiring an interested vascular segment from the two-dimensional coronary angiography image; picking a starting point and an ending point of the interested vascular segment; partitioning a local vascular area image corresponding to the starting point and the ending point from the two-dimensional coronary angiography image; filtering the local vascular area image to obtain a first image; performing a vascular enhancement on the first image to obtain a second image; extracting an initial vascular centerline from the second image; correcting the initial vascular centerline, and correcting a point deviating from a vascular center onto the vascular center to obtain an accurate vascular centerline.

Described herein are methods for cardiovascular imaging for diagnosing and/or detecting various cardiovascular diseases. Various embodiments of the invention provide using magnetic resonance imaging of the cardiovascular system of a subject at rest or a normocapnic condition, as well as at a stressed or hypercapnic condition, in a repeated manner enhancing the statistical power, such that fast, motion-corrected, free-breathing, whole-heart imaging of the cardiovascular system is utilized to identify impaired cardiovascular function in a manner with improved specificity and accuracy.

An ultrasonic diagnostic apparatus of an embodiment includes an ultrasonic probe and processing circuitry. The ultrasonic probe repeatedly performs scanning in which a plane wave or a diffused wave is continuously transmitted a plurality of times in the same direction in a plurality of directions. The processing circuitry performs processing of applying a moving target indicator (MTI) filter to an unequal interval data sequence in the same direction obtained by the scanning and extracting a blood flow signal in each of the plurality of directions, performs processing of generating an autocorrelation signal by performing an autocorrelation operation on a plurality of blood flow signals in the same direction for each of the directions, and estimates a velocity value of blood flow on the basis of a complex signal generated by performing complex addition of a plurality of autocorrelation signals generated for the plurality of directions.