Application of Machine Learning to Identify Obstructive Sleep Apnea Subgroups at Risk for Atherosclerosis Progression and Cardiovascular Disease Events (OSA-GRANDE)
Funded by NIH/NHLBI – R01HL168897-01A1
MPIs: Shah, Nadkarni, Suarez-Farinas 

The project aims to develop and validate predictive models using machine learning and artificial intelligence techniques to accurately identify individuals with obstructive sleep apnea who are at risk of experiencing subclinical cardiovascular progression, as well as primary and secondary cardiovascular disease events. 

Mount Sinai StARR Program
Funded by NIH/NHLBI – R38HL172261
MPIs: Kraft, Shah, Chu, Gallager 

The project aims to provide resident clinicians with a comprehensive curriculum focused on core competencies related to NHLBI-related disorders, offering coursework, laboratory rotations, and workshops for immersive research training. It also seeks to establish mentorship committees to guide individual career development and research projects contributing to the broader scientific literature. Furthermore, the project aims to support StARR residents in accelerating their research independence by providing career coaching, connecting them with fellowship programs in HLB diseases, and assisting in accessing subsequent postdoctoral awards such as T32, F32 NRSA support, individual StARR transition, and other individual K awards. 

Impact of Sleep Apnea in the Evolution of Acute Coronary Syndrome
Funded by AASM – 250-SR-21 
PI: Shah  

The project aims to investigate the relationship between obstructive sleep apnea (OSA) and cardiovascular disease (CVD) events, particularly focusing on the impact of continuous positive airway pressure (CPAP) therapy on these events. The researchers plan to use a machine learning approach to identify subgroups of patients within the ISAACC trial who may benefit from CPAP therapy or may experience potential harm. This approach is intended to address the previous lack of significant effects of CPAP treatment on CVD events, which may have been due to the heterogeneity of patient characteristics and the masking of treatment effects. 

Imaging the Atherosclerosis Cascade in Sleep Apnea 
Funded by NIH/NHLBI – R01HL143221
PI: Shah 

The project aims to address the gap in understanding the mechanistic link between obstructive sleep apnea (OSA) and atherosclerosis, particularly focusing on the role OSA plays in the progression of cardiovascular disease. By conducting a comprehensive assessment of the entire atherosclerosis cascade in OSA patients, including endothelial dysfunction, vascular stiffness, vascular inflammation, plaque composition, and total plaque burden, using state-of-the-art vascular imaging techniques, the study aims to unravel this relationship in detail. Additionally, the project seeks to investigate the anti-atherosclerotic actions of continuous positive airway pressure (CPAP) therapy, a common treatment for OSA, by identifying patients at high risk for atherosclerosis progression and those who respond most effectively to CPAP treatment. Through risk stratification of OSA patients and evaluating the effects of CPAP therapy on atherosclerosis within each risk stratum, the study aims to provide significant mechanistic insight and pave the way for more efficient, multi-site, randomized controlled trials to investigate the impact of OSA and CPAP therapy on cardiovascular event prevention. 

Genetic Epidemiology of Sleep Apnea and Comorbidities in Biobanks
Funded by NIH / Brigham and Women’s Hospital – R01HL153805
PI: Cade (Shah, Co-I) 

The project aims to conduct the most extensive genetic analysis to date of validated diagnosed cases of sleep apnea and insomnia. It intends to identify and characterize novel genetic loci associated with these conditions and examine their associations with clinical diagnosis data. By leveraging data from two large-scale biobanks (Partners and Geisinger), the project will collaborate with an interdisciplinary team of clinicians and bioinformaticians. The ultimate goal is to enhance patient classification and understanding of these sleep disorders, potentially leading to improved diagnostic and treatment approaches.