Mentoring and Research on Multimodal Data Integration to Predict Atrial Myopathy-Related Outcomes

About This Grant

This K24 grant renewal will continue to provide the PI, Dr. Lin Yee Chen, an NIH-funded patient-oriented research (POR) investigator, with the protected time and support needed to (1) accelerate his current trajectory in mentoring physicians and scientists who are conducting POR in cardiovascular disease (CVD), (2) acquire additional training in artificial intelligence (AI)/machine learning (ML) and omics science, (3) promote his current research that aims to use supervised ML and multimodal data integration to improve prediction of CVD and neurocognitive outcomes in people with atrial myopathy. Trainees in his AF/Atrial Myopathy Clinical Research Group will be recruited from the NIH-funded T32 training programs in the University of Minnesota's Division of Cardiology, Division of Epidemiology, and Division of Biostatistics; Department of Medicine Physician Scientist Training Program; K12 and T32 Programs of the CTSI; and graduate programs (MPH and PhD) in the School of Public Health. For his career development, Dr. Chen will hone his mentoring skills and learn new skills in cutting-edge areas (ML and omics science) through focused study, selected coursework, seminars, and guidance from senior collaborators with domain expertise. Finally, this grant will support a research project that is based on the Atherosclerosis Risk in Communities (ARIC) Study and Multi-Ethnic Study of Atherosclerosis (MESA), which extends Dr. Chen’s ongoing work to characterize the clinical importance of atrial myopathy. The specific aims are to develop and validate a multimodal prediction model for incident ischemic stroke (Aim 1), incident heart failure (Aim 2), and incident dementia (Aim 3) in participants with atrial myopathy by integrating a clinical risk score, polygenic risk score, ECG-based, echocardiogram- based, and proteome-based risk models. Models will be validated in the Cardiovascular Health Study (CHS). Our central hypothesis is that integrating multimodal data will improve prediction of clinical outcomes in participants with atrial myopathy compared to unimodal approaches. This project has significant impact: (1) This K24 renewal, which is focused on multimodal data integration and supervised ML, is a logical extension of the PI’s current K24 grant (K24HL155813) that is focused on unsupervised ML to classify atrial myopathy, (2) This K24 renewal will build upon the PI's exceptional mentorship track record. By assembling a team of senior collaborators that comprise experts in AI/ML, molecular epidemiology, and mentoring, the PI provides an outstanding platform for his mentees to acquire cutting-edge skills in POR, (3) By developing and validating comprehensive multimodal prediction models for CVD and dementia, this project will advance the NIH’s Precision Medicine initiative and NHLBI AI Initiative, (4) By integrating state-of-the-art data analytics and rigorous epidemiological methods with the rich resources of deeply phenotyped NHLBI cohort studies, this project will fill critical knowledge gaps in prevention and treatment, thus achieving a sustained and powerful impact on public health, clinical practice, and education of the next generation of researchers in POR.