Analyzed time-series data (Depressjon) to detect depression from patient activity recorded via clinical actigraphy watches. Utilized features such as time domain, statistical metrics, and LSTM-extracted attributes.
Done under the supervision of Dr. Anshika Arora & Ankur Maurya (Assistant Professors at Bennett University)