About | Sheida Nabavi | University of Connecticut

I am an Associate Professor in the School of Computing at the University of Connecticut (UConn). I received my PhD from Electrical and Computer Engineering Department at Carnegie Mellon University (CMU), where I also completed a postdoctoral fellowship. I subsequently worked as a Research Fellow and Research Associate at the Center for Biomedical Informatics (CBMI) at Harvard Medical School (HMS) before joining UConn in 2015, where I have established an interdisciplinary research program at the intersection of computing, biology, and medicine.

My research lies at the convergence of computational genomics, bioinformatics, and medical image analysis, with a focus on developing advanced machine learning and signal/image processing methods for analyzing complex biomedical data/images. My work emphasizes the integration of heterogeneous and multimodal data sources, including genomics, transcriptomics, histopathology, and medical imaging. A central theme of my research is the development of computational frameworks such as graph-based models, deep learning architectures, and probabilistic approaches, for capturing biological relationships and improve predictive modeling for disease diagnosis and outcomes.

I have contributed to the development of AI‑driven methods for medical imaging analysis, particularly in breast cancer detection using longitudinal mammography and digital breast tomosynthesis (DBT). My work introduces frameworks that model temporal changes across prior and current scans, enabling earlier and more accurate detection of subtle abnormalities. In parallel, my research in computational genomics advances methods for multi‑omics integration, biomarker discovery, and single‑cell and spatial transcriptomics analysis, providing insights into cellular heterogeneity and disease mechanisms. More recently, I have explored uncertainty‑aware and quantum‑inspired learning models, as well as continual and few‑shot learning techniques for adaptive clinical AI systems.