Big Data Applications to Biomedical Informatics



We use a variety of machine learning (ML) and data mining (DM) tools to provide insights into biomedical problems. This research is rooted in the explosion of data available for biomedical research. In recent years we have focused on temporal ML and DM as well as on understanding end-to-end uncertainty quantification for biomedical prediction and modeling. For the first, we are developing a comprehensive program in computational diabetes, with a focus on understanding the onset of T1D. For the second we are exploring both classical techniques such as Monte Carlo simulations as well recent advances in Conformal Prediction.

 Exemplar Publications

  • Mistry, S., Gouripeddi, R., Facelli, J.C. Data-driven identification of temporal glucose patterns in a large cohort of nondiabetic patients with COVID-19 using time-series clustering (2021) JAMIA Open 4:3 DOI: 10.1093/jamiaopen/ooab063.
  • Vazquez, J., Facelli, J.C. Conformal Prediction in Clinical Medical Sciences (2022) Journal of Healthcare Informatics Research.
  • Ye, X., Zeng, Q.T., Facelli, J.C., Brixner, D.I., Conway, M., Bray, B.E ; Predicting Optimal Hypertension Treatment Pathways Using Recurrent Neural Networks (2020) International Journal of Medical Informatics, 139, art. no. 104122. 
  • Pflieger, L.T., Pulst, S., Facelli, J.C. Characterization of Analytic and Experimental Uncertainty of RNA-seq Co-expression Network Determination: Application to SCA2 (2020) 2020 IEEE International Conference on Healthcare Informatics, ICHI 2020, art. no. 9374300.
  • Morid, M.A., Sheng, O.R.L., Fiol, G.D., Facelli, J.C., Bray, B.E., Abdelrahman, S ; Temporal pattern detection to predict adverse events in critical care: Case study with acute kidney injury (2020) Journal of Medical Internet Research, 22 (3), art. no. e1427. 
  •  Toggle Menu