Veer Sangha, an undergraduate researcher at Yale University, received the prestigious Elizabeth Barrett-Connor Research Award for Early Career Investigators for his presentation on the application of artificial intelligence (AI) to the electrocardiogram (EKGs). "Detection of Left Ventricular (LV) Systolic Dysfunction From Electrocardiographic Images” was published in the preprint server medRxiv.
Sangha is a member of the Cardiovascular Data Science (CarDS) Lab led by Rohan Khera, MBBS, MS. He was the first author on a study in Nature Communications on ECG Dx, a tool that uses AI to detect atrial fibrillation and other heart rhythm and conduction disorders.
The Elizabeth Barrett-Connor Research Award recognizes excellence in research by an investigator in training. Finalists are chosen from early career investigators whose abstracts are submitted to the American Heart Association’s Council on Epidemiology and Prevention (EPI). Sangha will present at the 2022 American Heart Association Scientific Sessions on November 5.
“This is an incredible accomplishment - as an undergraduate, Veer is among a very small group of trainees who have ever participated in this competition, which typically includes fellows and junior faculty. His selection is a testament to his rigor and creativity. Most importantly, his work has the promise to improve lives. We have a large population with cardiac dysfunction that is only diagnosed when symptoms emerge – losing out on the opportunity to reverse course with effective therapies during the early phase. However, cardiac imaging that is used to diagnose this is not practical for large scale screening,” said Khera.
The work pioneered by the CarDS Lab will allow the broad screening of LV systolic dysfunction directly from photos of ECGs. ECGs are routinely obtained in clinical settings globally. However, clinicians cannot determine from ECGs whether patients have LV systolic dysfunction. The deep learning-based tool allows broad use of accessible screening for patients and the use of ECG images, instead of digital signal data, can democratize access to advanced technology across various low-and high-resource settings globally.
Khera added, “We are committed to the further development of this technology – we are currently designing a large clinical study that prospectively tests the screening strategy, and are excited to be launching the SMART-LV study soon."