A new tool based in the deep learning model that has been developed in South Korea can be used to diagnose both the coronary areas and to predict large adverse heart events in an emergency.
How it works
Developed by a team of researchers from Yonsei University Severance Hospital, Keimyung University Dongsan Hospital and Medical Imaging AI Company Phantomics, automatically assesses the AI tool coronary CT angiography (CCTA) scans and classified stenose or OCCUSIVE.
The model also uses the YOLO architecture, which at the same time locates and classifies objects to quickly process images.
Findings
In a study, the AI model was tested using CCTA data from 408 patients who had acute chest pain at three emergency department from 2018 to 2022.
FindingsThey were published in the Radiology: Artificial Intelligence Journal of the Radiology Society of North America, noted that the in-depth analysis of the degree of stenosis was a better predictor of large adverse heart events (MACEs) than common clinical risk factors such as hyperlipidemia or the Hyperlipidemia.-T. T.
In addition, the linking of the AI-driven analysis with common risk factors improved the forecast of the channels by 14% points to 90%.
Why it matters
CT -angiography, which is used to assess arcadesosis for CAD prognosis, usually takes a long time to process results, where analyzes vary depending on the reader, according to Severance Hospital.
The AI tool that has been developed by the Korean research team not only detects CAD, but also predicts fuse risks in patients who occur in the Emergency Department.
“This study suggests the possibility that in -depth learning models can be applied to predict the patient’s prognosis, apart from simply determining the presence or absence of CAD in first auxiliary rooms, where rapid diagnosis and treatment decisions are important,” Dr. Jin Hur, professor at the Radiology department of Severance Hospital.
“The AI technology can be applied then simple diagnostic help to become a tool for clinical decisions,” he added.
Market Snapshot
Latest research projects in Asia-Pacific also used AI to improve the CAD diagnosis.
Singaporean startup Health Beta develops a solution that regards genetic and lifestyle factors in offering an improved polygene risk score for CAD. In the meantime, Three large heart hospitals in Singapore -Thet National Heart Center Singapore, the National University Hospital and Tan Tock Seng Hospital-being set to manage a new machine learning-powered system for fast CAD prediction.
In Australia, publicly mentioned medical devices Echo IQ and Artrya recently obtained 510 (K) approval from the United States Food and Drug Administration for their respective AI-driven software for diagnosing CAD. The product of ECHO IQ is specifically indicated for detecting serious aortic stome, while Artrya’s AI-driven software, Salix, delivers a 10-minute point-of-care assessment of CCTA scans.
#Korean #research #deep #learning #predicts #prognosis #coronary #artery #disease