Research Fellow

The petitioner proposes to develop advanced AI and machine learning models to improve diagnostic accuracy and disease prediction for neurological and oncological disorders. The work involves training AI systems on multimodal biomedical data, including genetic information and brain imaging, to identify early therapeutic targets. The goal is to enhance early detection and inform targeted therapy to improve patient outcomes and reduce U.S. healthcare costs.

The petitioner proposes to use natural language processing (NLP) and machine learning to extract insights from unstructured healthcare data, including clinical notes and medical images. He aims to develop deep learning frameworks and multimodal data fusion methods to help clinicians identify early-stage disease indicators and support treatment decision-making. The work focuses on refining model interpretability to reduce diagnostic errors and improve patient outcomes for chronic diseases like cancer.

The petitioner proposes to advance explainable machine learning and deep learning frameworks to create transparent, high-performing clinical models for the U.S. healthcare system. The work focuses on optimizing clinical workflows and improving diagnostic precision to support early disease detection and reduce healthcare costs. The endeavor aims to provide clinicians with actionable insights from complex data to ensure faster and more reliable medical decisions.

The petitioner proposes to develop state-of-the-art medical foundational models that utilize advanced continual learning and anomaly detection to detect distributional drifts in real-time. This work aims to ensure AI systems in healthcare avoid catastrophic forgetting while providing explainable and robust predictions for medical diagnostics.