Profile banner
Debayan Das

Debayan Das

Co-founder and clinical data scientist specializing in acoustic ML

dexter1618
Kolkata, India
Chatham, England, United Kingdom
Joined January 2026
Favicon imagedexter1618.github.io
Favicon imagegithub.com
Favicon imageorcid.org
Favicon imagescholar.google.com

Network

3.1K connections

Summary

Applied acoustic machine learning researcher focused on gait and voice biomarkers for older adults. His work centers on developing acoustic-based detection and triaging methods (gait triaging toolkit, footstep detection, voice-based triage for diabetes) that enable passive, non-intrusive monitoring in home environments. nih+2
Health‑tech practitioner and leader translating research into products and infrastructure. He holds leadership and founding roles (Interim Chief Data Officer at MiiCare; Co‑Founder at MiiHealth AI) and is responsible for data strategy, cloud architecture, security and production ML systems for older-adult care pathways. pydata+2
Active contributor to peer‑reviewed and conference research through collaborations. He has multiple conference papers and preprints (EMBC 2024, BSN 2023, arXiv papers) and participated in funded projects and awards related to healthy ageing and assistive AI. orcid+2
Early-career progression from intern to technical leadership with a hands-on engineering background. His personal portfolio documents progression from remote intern to full-time data scientist and subsequent leadership roles, alongside active open-source work and conference speaking. github+1

Work

Education

Projects

Writing

A Gait Triaging Toolkit for Overlapping Acoustic Events in Indoor Home Environments

July 1, 2024

Presents a machine-learning-based filter to triage gait audio samples from noisy indoor acoustic scenes, improving training data quality and gait detection performance for older adults in home environments.

Favicon imagepubmed.ncbi.nlm.nih.gov

A Voice-based Triage for Type 2 Diabetes using a Conversational Virtual Assistant in the Home Environment

January 1, 2024

Proposes a voice-based triaging system that extracts non-identifiable acoustic features from older adults' speech with a virtual assistant to pre-screen for Type 2 Diabetes, demonstrating feasibility for embedded voice assistant deployment.

Favicon imagearxiv.org

Passive tracking of gait biomarkers in older adults: Feasibility of an acoustic based approach for non-intrusive gait analysis

January 1, 2023

Conference paper exploring acoustic gait analysis for at-risk community-dwelling older adults (conference publication / peer-reviewed proceedings).

Favicon imagescholar.google.com

Hobbies

Open-source coding and software development (active GitHub contributions). github+1