Opportunistic use of existing chest CT data for frailty screening - PROJECT SUMMARY I am an early-stage clinician investigator and an advanced multi-modality imaging cardiologist at the Icahn School of Medicine at Mount Sinai. My goal is to become a leader at the intersection of aging, cardiovascular imaging, and cardiovascular disease. My work focuses on how best to deliver patient-centered care and improve outcomes of older adults through use of cardiovascular imaging for stable ischemic heart disease (SIHD). Assessment of frailty is one key tool to improve these outcomes, as one in four of the 10 million older adults in the U.S. with SIHD is frail. Frailty is associated with increased risk of complications, hospitalizations, health care costs and mortality among patients with or without SIHD. Unlike most geriatric syndromes, frailty can be reversed with targeted interventions if diagnosed early and can guide personalized treatment decisions. Yet, frailty is rarely assessed in clinical practice due limited time and resources, or unfamiliarity with screening tools. This may lead to clinicians withholding appropriate beneficial treatment options for certain patients or subject other patients to inappropriate treatments that increase harm with limited benefit. Novel frailty screening tools that overcome these barriers are critical to increase frailty screening in clinical practice. The opportunistic use of existing imaging data acquired for other clinical indications is an innovative application for frailty measurement. These imaging data are ubiquitous: majority of the more than 1.5 million stress nuclear myocardial perfusion imaging (MPI) studies performed annually in older adults for diagnosis and management of SIHD also acquire low-dose chest CT images to improve image quality and diagnostic performance. In this project, I will develop and validate a novel chest CT-based frailty screening tool using artificial intelligence (AI) assisted multi-organ body composition information available from existing chest CT data performed with stress MPI. Next, we will compare the prognostic performance of the new CT-based tool with a validated electronic health record-based electronic frailty index to predict healthcare utilization and death. Finally, we will adapt this screening tool for integration within routine clinical workflow, and pilot test its feasibility using a pre-post approach among older adults presenting to an outpatient cardiology clinic visit for clinical decision making after stress imaging. The results of our study will integrate novel frailty screening into routine cardiovascular practice, to help guide patient-centered treatment decisions, target effective prevention strategies, and inform my future R01, which will test AI-enhanced patient evaluation strategies among older adults. This Beeson award will support my training in (1) geriatrics and geriatric cardiology; (2) clinical applications of AI (3) dissemination and implementation science and (4) leadership. Our approach can expand the availability of real-time point-of-care frailty screening without additional clinical burden by using existing CT data. The Beeson award will help me achieve my goal to become a leader in aging who develops novel strategies aligning imaging use with the distinct needs of older adults with SIHD.
