Methods, devices and media for determining flow information of coronary arteries based on coronary CT angiography images of patients

Abstract

The present disclosure relates to a method, device and medium for determining flow information of coronary arteries based on coronary CT angiographic images of a patient. The method includes: acquiring a CCTA image of the patient; extracting arterial features of the blood vessel of interest based on the acquired CCTA image; extracting corresponding myocardial features of the blood vessel of interest based on the acquired CCTA image; The features and at least a subset of corresponding myocardial features are used to predict flow information of the coronary arteries using a trained prediction model. This method can obtain more accurate and patient-specific coronary artery flow information from CCTA images, and can further obtain more reliable virtual FFR predictions based on this, and can also provide doctors with quantitative patient-specific flow physiology information from routine CCTA examinations .

Description

[0001] Cross References to Related Applications [0002] This application claims priority to U.S. Provisional Application No. 62 / 726,989, filed September 5, 2018, the entire contents of which are incorporated herein by reference. technical field [0003] The present disclosure generally relates to medical image processing and analysis. More specifically, the present disclosure relates to methods, devices and media for determining flow information of coronary arteries based on medical images of coronary arteries of a patient. Background technique [0004] Typically, the primary diagnostic method for coronary artery disease relies on the invasive measurement of fractional flow reserve (FFR). Recently, non-invasive methods such as computational fluid dynamics and deep learning techniques have been applied to coronary computed tomography angiography (CCTA) images to derive FFR (also called virtual FFR). The flow information of the patient's coronary arteries, such as inlet flo...

AI Technical Summary

Problems solved by technology

However, this intermediate information is usually estimated using empirical relationships obtained from simple theoretical equations or statistical correlations, resulting in estimated flow information in the patient's coronary arteries that is not patient-specific (i.e., inaccurately related to the patient's true condition). match), on this basis, the virtual FFR is obtained, which significantly affects the prediction accuracy of the virtual FFR

Images