Research
My work sits at the intersection of AI, deep learning, and healthcare — building interpretable, trustworthy systems for real clinical impact.
Medical Imaging
Segmentation, detection, and reconstruction in MRI/CT with clinical constraints. Current focus: brain tumor segmentation with uncertainty estimation.
Explainable AI
Human-centered explanations for deep learning models. Developing XAI methods that reduce cognitive load for clinicians in high-stakes settings.
Health Analytics
Population-level modeling, risk prediction, fairness and bias analysis in AI health pipelines. Parkinson's severity from pen signals.
Large Language Models
Exploring LLMs in clinical text understanding, biomedical NLP, and knowledge extraction for healthcare decision support.
Computer Vision
Object detection, image classification, and video understanding applied to medical and real-world scenarios.
AI Fairness & Ethics
Auditing AI systems for bias in health contexts; developing equitable models across demographic groups.
Featured Projects
Explainable MRI Tumor Segmentation
Interpretable deep learning model for brain tumor segmentation using uncertainty estimation and attention-based explanations. Designed to align with clinician workflows and reduce diagnostic uncertainty.
Parkinson's Severity Detection from Pen Signals
Learning motor impairment signatures from handwritten spiral and wave drawings. Non-invasive, low-cost screening system validated against clinical ratings.
Trustworthy XAI for Clinical Decision Support
Building clinician-aligned explanation frameworks that reduce cognitive load in high-stakes medical decisions. Evaluated through user studies with healthcare professionals.