Headline

Researcher in Computational Neuroscience | EEG Analysis | Neurodiagnostics | Specializing in Statistical Modeling & Advanced Machine Learning.

About

🧠 Can a machine learn to read your brain better than your doctor can? I'm spending my PhD finding out.......... I work at the intersection of EEG signal processing, machine learning, and computational neuroscience, building intelligent systems that decode brain activity to detect neurological disorders like Autism Spectrum Disorder (ASD) and Epilepsy. My goal is simple: turn complex, noisy brain data into tools that actually help clinicians make better decisions, faster. My research combines signal processing, explainable AI, and deep learning on real-world, large-scale EEG datasets. 🔬 What I built so far: • XAI pipelines ( SHAP) for EEG microstate-based ASD diagnosis. • CNN-LSTM models for multi-year seizure forecasting. • Stochastic modelling of brain microstate dynamics for biomarker discovery. 🛠 Tools & Skills: Python · PyTorch · TensorFlow · Scikit-learn · MNE-Python · EEGLAB · SHAP · CNN · LSTM · MATLAB Also I am actively looking for research internship opportunities in health AI, computational neuroscience, and biomedical ML. If your work lives somewhere between the brain and the algorithm, then let's talk. 📧 delna.kuriyakose@vit.ac.in

Location

Vellore

Work experience

Research Scholar at Vellore Institute of Technology

Researcher at Collaboration with Dr. Radhakrishnan Delhibabu, Computational Neuroscience, 2025 - Present · 1 yr 6 mos, Collaborated on deep learning approaches for epilepsy analysis including seizure forecasting and temporal EEG stability studies, resulting in two manuscripts currently under review in Q1/Q2 journals. Joint research on developing clinical alarm validation systems for seizure prediction with translational neurology applications.

Doctoral Researcher at Department of Mathematics, VIT Vellore, Sep 2023 - Present · 2 yrs 10 mos, Developed an explainable AI pipeline (XGBoost + SHAP) integrating EEG microstate dynamics for ASD classification, achieving state-of-the-art diagnostic accuracy on benchmark datasets. Designed a stacked ensemble framework combining spectral, microstate, and connectivity EEG features for robust neurodiagnostic assessment of Autism Spectrum Disorder. Built a CNN-LSTM regression framework for longitudinal seizure Remaining Useful Life (RUL) forecasting. Applied stochastic and Markov-chain modelling to characterise EEG microstate transition dynamics as neurophysiological biomarkers for ASD.

Education

Doctor of Philosophy, Mathematics, Vellore Institute of Technology

Master's degree, Statistics, University of calicut