Assistant Professor (Tenure Track), Department of Computer Science, University of South Dakota

Research Overview: My primary research focuses on developing advanced artificial intelligence algorithms to improve medical imaging across clinical domains including upper and lower gastrointestinal imaging, lung and liver tumor analysis, and predictive modeling for radiation therapy outcomes. Diagnostic accuracy is often limited by data scarcity, interobserver variability, biases, and limited generalizability. To address these issues, I curate multinational datasets including CirrMRI600+ for liver cirrhosis imaging, PolypDB, PolypGen, and Kvasir-SEG for colonoscopy, as well as HyperKvasir and KvasirCapsule for GI endoscopy and video capsule endoscopy. I also develop segmentation architectures such as ResUNet++, DoubleUNet, ColonSegNet, and transformer-based models that enhance diagnostic accuracy. In prior work I developed algorithms for colonoscopy and endoscopy. One of our algorithms, ColonSegNet, and the dataset Kvasir-SEG have been included by NVIDIA Clara. I investigate predictive modeling and organ-at-risk assessment to optimize radiation therapy planning and outcomes. By integrating curated datasets with state-of-the-art deep learning, my research aims to improve the precision, generalizability, and clinical effectiveness of AI-driven diagnostics and therapies in healthcare.