Bhanu Teja Gullapalli

Bhanu Teja Gullapalli is a Postdoctoral Fellow at Harvard University's John A. Paulson School of Engineering and Applied Sciences, collaborating with Susan Murphy. His research bridges wearable health sensing and machine learning to address substance use disorders through digital biomarkers.

Education

• PhD in Data Science, University of California San Diego (2024)
• MS/PhD in Computer Science, University of Massachusetts Amherst (2017-2022)
• Bachelor's degree, Indian Institute of Technology Guwahati

Gullapalli's work focuses on extracting clinically relevant insights from multimodal physiological data (e.g., heart rate variability, PPG signals) collected via wearables. He develops machine learning models to predict opioid/cocaine craving, euphoria, and administration moments, integrating pharmacokinetics and behavioral therapy principles. His research enables just-in-time adaptive interventions for addiction treatment, emphasizing real-world applicability in both clinical and naturalistic settings.

His 7 publications (2019-2024) in venues like EMBC and npj Digital Medicine demonstrate consistent innovation in wearable-based addiction monitoring. Key trends include advancing temporal modeling for substance use prediction, leveraging earbud-based PPG for stress detection, and creating closed-loop systems that trigger mindfulness interventions during craving episodes.

Scientific Recognition

• Future Leaders Summit, Michigan Institute for Data Science (2023)
• Innovation to Impact program, Yale University (2021)

Gullapalli contributed to an NSF Smart and Connected Health grant ($1.1M) supporting wearable-based SUD research at UMass Amherst. He actively collaborates with medical researchers including Eric L. Garland (University of Utah) and industry partners at Optum AI Labs and Samsung Digital Health Lab to translate research into clinical tools.

His work occurs within Susan Murphy's Harvard lab and cross-institutional teams focused on mobile health interventions, with ongoing projects developing earbud-based stress detection systems and pharmacokinetics-informed neural networks for opioid use prediction.