Rohan Khera

Dr. Rohan Khera is a Cardiologist-Data Scientist at Yale University and leads the Cardiovascular Data Science (CarDS) Lab. The goal of his group is to develop and implement strategies to improve outcomes for patients with or at risk for cardiovascular disease through data-driven innovations in delivering evidence-based, patient-centered care. His work focuses on novel strategies for learning from complex clinical data and incorporating information and technology in healthcare to improve care efficiency and provide actionable insights to improve patient outcomes. The ongoing work includes, (1) data science innovation that leverages structured and unstructured elements in the electronic health record (EHR) to evaluate quality of care and their association with patient outcomes, (2) applications of machine learning to achieve precision inference from clinical trials, (3) deep learning and artificial intelligence to enhance novel disease detection from wearable devices, electrocardiography, cardiac imaging, and natural language, and (4) methodological investigations focusing on improving the rigor of studies that use large datasets.

The CarDS Lab has a strong focus on mentorship and career development and currently supports research trainees across the career spectrum, including students in data science, physician-scientists-in-training, graduate students, and clinical trainees. The work in the Lab is supported by grants from the National Institutes of Health, Doris Duke Charitable Foundation, industry, and Yale University, offering diverse experiences. The goal of the program is to be at the forefront of innovation in health data science and to train the next generation of leaders in the field.

What do you do with data science?

“The group has a strong focus on innovation in healthcare data science with work focusing on digital phenotyping of cardiovascular disease and the development of automated assays of care quality within the electronic health record. Key areas of investigation include (1) data science innovation that leverages structured and unstructured elements in the electronic health record (EHR) to evaluate quality of care and their association with patient outcomes, (2) applications of machine learning to achieve precision inference from clinical trials, (3) deep learning and artificial intelligence to enhance novel disease detection from wearable devices, electrocardiography, cardiac imaging, and natural language, and (4) methodological investigations focusing on improving the rigor of studies that use large datasets.

Some recent publications that highlight the focus of our team:

a) Sangha V, Mortazavi BJ, Haimovich AD, Ribeiro AH, Brandt CA, Jacoby DL, Schulz WL, Krumholz HM, Ribeiro ALP, Khera R. Automated multilabel diagnosis on electrocardiographic images and signals. Nature Communications 2022,13:1583. PMID: 35332137

b) Oikonomou EK, Van Dijk D, Parise H, Suchard MA, de Lemos J, Antoniades C, Velazquez EJ, Miller EJ, Khera R. A phenomapping-derived tool to personalize the selection of anatomical vs. functional testing in evaluating chest pain (ASSIST). European Heart Journal. 2021 Apr 21: ehab223. PMID: 33881513.

c) Khera R, Haimovich J, Hurley NC, McNamara R, Spertus JA, Desai N, Rumsfeld JS, Masoudi FA, Huang C, Normand SL, Mortazavi BJ, Krumholz HM. Use of Machine Learning Models to Predict Death After Acute Myocardial Infarction. JAMA Cardiology. 2021 Mar 10:e210122. PMID: 33688915.

d) Khera R, Mortazavi BJ, Sangha V, Warner F, Patrick Young H, Ross JS, Shah ND, Theel ES, Jenkinson WG, Knepper C, Wang K, Peaper D, Martinello RA, Brandt CA, Lin Z, Ko AI, Krumholz HM, Pollock BD, Schulz WL. A multicenter evaluation of computable phenotyping approaches for SARS-CoV-2 infection and COVID-19 hospitalizations. npj Digital Medicine. Nature Publishing Group; 2022 Mar 8;5(1):1–9.

e) Holste G, Oikonomou EK, Mortazavi BJ, Faridi KF, Miller EJ, Forrest JK, McNamara RL, Krumholz HM, Wang Z, Khera R. Automated detection of severe aortic stenosis using single-view echocardiography: A self-supervised ensemble learning approach. medRxiv. 2022. doi: 10.1101/2022.08.30.22279413