Research and academic projects across clinical AI, medical imaging, neuroengineering, and biomedical NLP.
Master's thesis project: multi-tier evaluation framework to validate autonomous AI agents for NSCLC treatment planning. Combines LLM-as-judge, deterministic rule checks, and clinician feedback loops to ensure safety, accuracy, and alignment with clinical guidelines.
View on GitHubDeep learning pipeline to decode and classify perceived visual stimulus categories from human brain activation maps using fMRI data. Conducted ROI-based feature analysis and architectural exploration across CNN, RNN, and fully connected networks, with explainability techniques to evaluate voxel-level contributions.
View on GitHubDeep CNN pipeline for automated multi-class classification of 96x96 high-resolution white blood cell images, plus a semantic computer vision segmentation architecture for identifying structural features in planetary terrain images.
View on GitHubTrained image classifiers using pre-trained network architectures to differentiate between benign and malignant pathology features. Established confidence-interval estimations with a dual multi-input processing pipeline utilizing full-slice and zoomed-slice inputs.
View on GitHubDeveloped an interactive conversational AI medical question-answering assistant fine-tuned on the PubMed Q&A biomedical domain dataset.
View on GitHubCoded a 3D multimodal MRI semantic parsing algorithm to execute automated brain ventricle volume segmentation mappings.
Code on requestDeveloped a physiological signal processing pipeline to extract wave characteristics from raw data streams and classified discrete sleep macro-architectures including REM, NREM, and awake stages based on EEG variations.
Code on requestBuilt localized baseline data classifiers to predict corporate loan approval criteria and environmental pollution metrics. Constructed predictive optimization algorithms for urban environmental telemetry data to simulate flood risks.
View on GitHub