IEEE URTC
Linguistic Biomarkers for Dementia Detection
- Accepted to the IEEE MIT Undergraduate Research Technology Conference (URTC).
- Developed models using linguistic features to predict dementia.
- Compared multiple machine learning approaches and evaluated performance.
- Demonstrates experience at the intersection of NLP, healthcare, and AI.
IEEE AIxSET
Boosting vs. Multiclass Models in Parkinson’s Detection
- Accepted to the IEEE Conference on AI, Science, Engineering, and Technology (AIxSET).
- Designed and benchmarked ML pipelines to classify Parkinson’s disease.
- Compared boosting-based models with multiclass approaches.
- Highlights ability to design, implement, and interpret ML systems in healthcare contexts.
TechRxiv — Preprint Server
Ensemble Learning for Aphasia Diagnosis via Vocal Biomarkers
- Introduced a machine learning stacking ensemble to objectively diagnose aphasia and assess severity.
- Reduced reliance on subjective clinical judgment through data-driven modeling.
- Combined multiple ML classifiers to achieve more reliable and robust predictions.
- Demonstrated the applicability of AI in neurological and speech-language disorder assessment.
- Published as a TechRxiv preprint, contributing to interdisciplinary research in AI and healthcare.
Impact
Research Readiness
- Together, these two publications highlight Nandan's advanced preparation in AI/ML research.
- Shows experience in ideation, experimentation, coding, analysis, and academic writing.