Saint Louis University (SLU)
My research bridges data-centric AI and clinical applications by developing machine learning methods to analyze high-dimensional, multimodal health data. I focus on missing data imputation, change-point detection, and unsupervised representation learning using generative models, such as GANs and diffusion models. My work includes modeling disease progression using digital biomarkers and integrating CNS biomarker data to stratify neurological risk. I also apply AI for pharmacokinetic prediction and drug safety modeling. My previous experience also involves analyzing single-cell RNA-seq data to understand biological heterogeneity and cell-type-specific expression patterns across disease stages. I also apply machine learning to estimate pharmacokinetic parameters and drug exposure, with a growing interest in supervised mixed-membership models and constrained matrix factorization for biomarker-driven clustering. Through these methods, I aim to advance precision health by enabling robust, interpretable, and scalable AI systems that improve early detection, monitoring, and intervention strategies in aging and cognitively vulnerable populations.