AI Developer × Bioengineer
Pushing the boundaries of artificial intelligence and bioengineering to create innovative solutions that bridge technology and medicine.
I'm a PhD researcher specializing in the intersection of artificial intelligence and biomedical engineering. My work focuses on developing multi-modal AI systems for early disease detection, particularly in Parkinson's Disease screening using advanced machine learning techniques.
With expertise in deep learning, computer vision, signal processing, and full-stack development, I build end-to-end AI solutions that make a real impact in healthcare and beyond.
Hybrid Vision Transformer + KNN model for Parkinson's Disease detection from spiral drawings. A key architectural insight: augmentation strategies must be architecture-aware. The model performs best on original, unaugmented data, challenging common assumptions in medical AI.
Unified deep learning ensemble framework combining multimodal fusion with stacked tree-based models for voice-based PD detection and motor severity prediction (UPDRS). Achieves R² of 99.78% on regression with RMSE of 0.3802, validated via 10-fold CV.
A second brain for academic researchers. Auto-detects and saves papers from 10+ publishers (arXiv, PubMed, Nature, IEEE…), manages submission deadlines, and uses an AI-powered recommendation engine to surface the most relevant journals and conferences based on your research profile. 100% local. Zero servers, zero telemetry.