Duke Plus Data Science Program

Project goal

Identified the drivers of the risk of coronary heart disease and cardiovascular disease using the Sleep Heart Health Study (SHHS) dataset.

Details

1. Conducted data pre-processing including transformations, imputation, deduplication, and construction of survival labels.
2. Performed Exploratory Data Analyses using paired t-test and factor analysis to quantify univariate associations and identify underlying structures of the data.
3. Trained the Lasso-Cox Proportional Hazards model and XGBoost-Accelerated Failure Time model to measure the contributions of factors towards increasing or decreasing risk and to model complex feature interactions.

Software

R - 4.2.2

Data reference

1. Zhang GQ, Cui L, Mueller R, Tao S, Kim M, Rueschman M, Mariani S, Mobley D, Redline S. The National Sleep Research Resource: towards a sleep data commons. J Am Med Inform Assoc. 2018 Oct 1;25(10):1351-1358. doi: 10.1093/jamia/ocy064. PMID: 29860441; PMCID: PMC6188513.
2. Quan SF, Howard BV, Iber C, Kiley JP, Nieto FJ, O'Connor GT, Rapoport DM, Redline S, Robbins J, Samet JM, Wahl PW. The Sleep Heart Health Study: design, rationale, and methods. Sleep. 1997 Dec;20(12):1077-85. PMID: 9493915.

Acknowledgement

The Sleep Heart Health Study (SHHS) was supported by National Heart, Lung, and Blood Institute cooperative agreements U01HL53916 (University of California, Davis), U01HL53931 (New York University), U01HL53934 (University of Minnesota), U01HL53937 and U01HL64360 (Johns Hopkins University), U01HL53938 (University of Arizona), U01HL53940 (University of Washington), U01HL53941 (Boston University), and U01HL63463 (Case Western Reserve University). The National Sleep Research Resource was supported by the National Heart, Lung, and Blood Institute (R24 HL114473, 75N92019R002).