MCIRCC Data Science Team Director Awarded NIH Grant to Study Applications of Machine Learning, Signal Processing, and Wearable Technologies to Predict Heart Failure

Heart failure is one of the leading causes of death in the United States, as well as a major driver of healthcare costs. The American Heart Association predicts that by the year 2030 more than 8 million people will be living with heart failure, with projected direct costs doubling to $53 billion nationwide*. 

Sardar Ansari, PhD Director, MCIRCC Data Science Team Research Assistant Professor, Emergency Medicine

Sardar Ansari, PhD
Director, MCIRCC Data Science Team
Research Assistant Professor, Emergency Medicine

Currently, it is difficult to predict who will develop heart failure, but one MCIRCC researcher seeks to bring us closer to doing so. Sardar Ansari, PhD, the director of MCIRCC’s Data Science Team and a Research Assistant Professor in the Department of Emergency Medicine, has been awarded a five-year K01 grant from the National Institutes of Health (NIH) to study the applications of machine learning and signal processing techniques on multiple types of data, including data collected by consumer wearable devices, to predict the onset of heart failure.

The data types that Dr. Ansari is focusing on are electronic health records (EHR), electrocardiograms (ECG), and heart rate variability (HRV). “EHRs contain information that can be used to predict heart failure before its onset, yet currently existing models lead to a large number of false positive predictions,” he notes. As such, a major aim of his study is to improve the accuracy of these predictions by augmenting EHR data with ECG and HRV features. These three modalities of data will be analyzed using novel deep learning methods, and the models will be developed and validated retrospectively using patient data available at Michigan Medicine.

Dr. Ansari also aims to expand the impact of this research to larger populations by replacing clinically measured ECG and HRV values with those obtained by consumer wearables such as smart watches. As part of the study, a prospective patient cohort will use a wearable device that collects intermittent ECG and continuous HRV, allowing Dr. Ansari to determine whether this data, when combined with EHR, can provide clinicians with a more effective tool to identify which patients are at risk of heart failure. He will also evaluate a limited model that relies only on the information gathered by the wearable device in order to include patient populations who don’t have a past medical history. 

“We expect the resulting tools will help prevent delayed diagnosis of heart failure, leading to improved medical outcomes, enhanced patient experience, and reduction in healthcare costs,” said Dr. Ansari.

The NIH K01 Mentored Research Scientist Career Development Award provides support and protected time for an intensive, supervised career development experience in the biomedical sciences. The award will allow Dr. Ansari to acquire needed additional training in cardiovascular physiology and heart failure pathophysiology. Dr. Ansari’s mentoring team is led by Drs. Kevin Ward and Keith Aaronson. 

References

Kate MurphyFeatured Research Project