We are delighted to welcome Srinivasan Vairavan , Director of Digital Health and AI Biomarkers at Johnson & Johnson Innovative Medicine R&D and Visiting Lecturer at King's College London , to deliver his seminar, "Precision Psychiatry through Digital Phenotyping: Lessons Learned from Three Longitudinal Studies in major depressive disorder" , as part of the EPSRC DRIVE-Health Seminar Series. Abstract: Major Depressive Disorder (MDD) remains diagnosed and monitored almost entirely through episodic, subjective self-report, limiting both clinical care and trial sensitivity. This lecture presents evidence that passively collected digital signals can objectively characterize core MDD symptom domains and track an individual's mental state continuously between clinic visits. Drawing on more than 1,000 patients across three major longitudinal cohorts — RADAR-CNS, ORBIT (NCT02489305), and CANBIND — we describe methods that translate subtle behavioral and physiological data into clinically interpretable biomarkers. Paralinguistic analysis of speech is used to quantify psychomotor retardation, while fractal signatures derived from actigraphy index depression severity and impending relapse. Building on these markers, we present a personalized, N-of-1 anomaly detection framework that fuses passive digital streams with intermittent self-report to flag elevated relapse risk at the individual level. Results support the clinical validity of this approach. Digital measures, including speech intensity, showed significant associations with clinical outcomes. The relapse prediction model achieved a balanced accuracy exceeding 71%, with a median detection lead time of two to three weeks prior to clinical onset and a low false alarm rate — a profile compatible with real-world deployment. The lecture will also address principal sources of measurement variability, including seasonal effects, and the emerging relationship between cognitive performance and digital biomarkers. Together, these findings indicate a shift from episodic subjective assessment toward objective, unobtrusive, and continuous monitoring in mental healthcare — with near-term implications for patient selection and endpoint sensitivity in clinical trials, and longer-term potential for personalized diagnosis and treatment in MDD. Seminar Series Event : "Precision Psychiatry through Digital Phenotyping: Lessons Learned from Three Longitudinal Studies in major depressive disorder" Date and Time: Thursday 15 October 2026, 13:30 – 14:30 (BST) Location: The River Room, King’s Building, 2nd Floor, Room KIN 227, Strand Campus Attendance: Mandatory for all DRIVE-Health students; a calendar invitation has already been sent. Registration: Alumni and the wider King's College London research community are welcome - no booking required. Biography Srinivasan Vairavan , PhD, is Director of Digital Health and AI Biomarkers within Neuroscience Data Science and Digital Health at Johnson & Johnson Innovative Medicine R&D, and a Visiting Lecturer in the Department of Biostatistics and Health Informatics at the Institute of Psychiatry, Psychology and Neuroscience, King's College London. He has over 15 years of experience building and scaling AI, machine learning and advanced analytics capabilities across pharma and healthcare, with prior roles at Proteus Digital Health and Philips Research. His work translates large-scale clinical, digital health, molecular and EHR data into regulatory-grade endpoints, including engagement through FDA Type C meetings and EMA Qualification Advice, and co-development of the industry-wide V3+ framework. He leads and contributes to major international consortia including RADAR-CNS, RADAR-AD and AMP-SCZ. He holds patents in relapse prediction in major depressive disorder and speech-based detection of cognitive decline, sits on the editorial board of BMC Digital Health, and serves on the DRIVE-Health advisory board.