Enabling startup, clinicians and researchers to translate biomedical data into actionable insight through interpretable ML and generative AI.
I help clinicians researchers & digital health teams to:
- ๐ฌ Enable clinicians and researchers to apply state-of-the-art AI to biomedical signals through intuitive, no-code interfaces
- ๐ง Design interpretable & publishable ML pipelines for translational research
- ๐ Deploy scalable end-to-end solutions in clinical or startup environments
- Prodromal_Parkinson_XAI โ Predictive modeling and SHAP explainability for Parkinson prodrome detection from IMU data
- EMG_HSP_XAI โ Deep learning + SHAP on EMG signals to identify muscle drivers in spastic gait (HSP disorder)
- FallRiskPredictor โ Explainable Streamlit webapp to estimate fall risk in neurological patients
- Fall risk prediction with XAI for Parkinsonโs patients
- Prodromal signature discovery using SHAPSetPlot on wearable data
- Muscle importance analysis for rare disorders via BiLSTM + CNN on EMG
- Synthetic data generation (ctGAN) for class balancing in clinical datasets
- Machine Learning Approach to Support the Detection of Parkinson's Disease via IMU Gait Analysis
Published in Sensors (2022) โ 100+ citations - Optimizing Rare Disease Gait Classification through Data Balancing and Generative AI
Published in Sensors (2024) โ 20+ citations
I'm open to research, clinical or AI product collaborations in:
- Prodromal Parkinson detection
- Generative AI for rare disease datasets
- Clinical explainability (SHAP / SHAPSetPlot)
- Fall prediction systems
๐ฉ Letโs connect if you'd like to collaborate!
