MORGANTOWN, W.Va. – Recent editorials written by Partho Sengupta, M.D., chief of Cardiology and chair of the Center of Innovation, and Sirish Shrestha, M.Sc., biostatistician and machine learning research scientist, at the WVU Heart and Vascular Institute provide framework regarding the use of artificial intelligence for precise and early detection of cardiac disease.
According to the editorials, including one in the April issue of “Circulation: Cardiovascular Imaging,” a publication of the American Heart Association, computer modeling and algorithms in machine learning can provide opportunities to resolve difficulties in diagnosis and minimize impediments that arise from high-dimensional imaging datasets. Machine learning, a subset of artificial intelligence, can leverage large amounts of data and automated reasoning to detect and discern patterns from a large number of patients to help predict health outcomes.
These editorials follow the recent publication by Dr. Sengupta and his research partners in JACC: Cardiovascular Imaging, titled “Phenotypic Clustering of Left Ventricular Diastolic Function Parameters,” which examined how machine learning can be used to better diagnose left ventricular diastolic dysfunction and how it outperformed currently existing guideline standards for predicting cardiac mortality and hospitalizations. Sengupta and his colleagues started this study at Mount Sinai in New York, later including patient data from West Virginia after he joined WVU Medicine.
“These are all firsts for West Virginia and WVU Medicine,” Sengupta said. “It’s the coming together of research scholars, engineers, and study personnel as a team that enables such rapid advances. There is nothing more satisfying than to see the team succeed, and it’s wonderful to see fellow researchers from Europe, Japan, and Korea who are joining the team to discover new knowledge that accelerates patient care.”
In the year since the Innovation Center began, the WVU Heart and Vascular Institute has had repeated successes in publishing innovative research on the use of machine learning in cardiology.
“We have attracted engineers and professionals from across the world who have come together to create a network of enthusiastic people with complimentary roles,” Sengupta said. “There are young clinical researchers who have come and joined as research scholars.”
Marton Tokodi, a cardiovascular research scholar who joined the WVU Heart and Vascular Institute from Hungary, has been named a finalist in the 2018 Arthur E. Weyman Young Investigator’s Award competition, which is supported by the National Board of Echocardiography. He will have the opportunity to present his research “Extracting Knowledge from Geometric Shape of Echocardiography Data” at the Annual Scientific Sessions of the American Society of Echocardiography in Nashville, Tenn., and, if he is selected as the overall winner, he will also be invited to present at the December 2018 European Association of Cardiovascular Imaging’s EuroEcho-Imaging meeting and the May 2019 Japanese Society of Echocardiography’s Annual Scientific Meeting.
Additionally, Sengupta and his team at the Center have been invited by the “Journal of the American College of Cardiology” to develop a consensus document on machine learning to serve as a roadmap for innovation in cardiac imaging.
“This all shows how we are emerging as leaders in cardiac imaging and in new technologies,” Sengupta said. “We are enabling people from different disciplines to be successful cardiologic researchers.”