Title: Modeling, Monitoring & Modulating Biobehavioral State Dynamics for Just-in-Time Health Support
Asim H. Gazi, PhD, Schmidt Science Fellow, Harvard University
Abstract: A patient engaged in therapy for one hour a week spends 99% of their time without clinical support. Mobile health (mHealth) technologies aim to address this gap by integrating biobehavioral monitoring and intervention in daily life. Yet, key technical challenges remain that preclude this aspiration from materializing. In this talk, I will discuss how we have addressed some of these challenges when modeling, monitoring, and modulating (non-invasively) the autonomic nervous system for patients with posttraumatic stress disorder reliving traumatic memories. These challenges include the estimation of clinically meaningful "physio-markers" from raw sensor data in the presence of in-band noise and the modeling of physiomarker dynamics in response to nVNS at the timescale necessary to assess potential efficacy in a "just-in-time" setting. Central to our approach is the thesis that methods informed by domain science (e.g., psychophysiology) outperform domain-agnostic methods in health settings where data are often limited and the need for physician scientists' trust is critical. Carefully considering these factors, we designed signal quality assessment and physio-marker extraction techniques informed by cardiovascular and respiratory physiology and demonstrate improved quantification of physiological changes relevant to stress. We also show that by investigating physiomarker changes in response to nVNS and stressors through a dynamical modeling lens, we unearth insights pertinent to domain scientists. For example, our research is the first to show that nVNS can counteract some of the physiological effects of trauma recall-induced stress within 10-15 seconds. Through collaborative efforts, this work is now stepping toward clinical impact. Looking ahead, I will conclude the talk with an overview of current research efforts, specifically focusing on collaborative research underway with Inki Kim and team at UIUC.
Bio: Asim H. Gazi is a Schmidt Science Fellow at Harvard University, advised by Susan A. Murphy. He completed his B.S. in electrical engineering from the University of Texas at Dallas in 2018 and his Ph.D. in electrical engineering from the Georgia Institute of Technology (GT) in 2023. His PhD was advised by Omer T. Inan and Christopher J. Rozell and was funded by the National Science Foundation Graduate Research Fellowship. Asim's interdisciplinary research is on the design of models and algorithms to help tailor healthcare support to changes in a patient's physiology or behavior. His methods are informed by psychophysiology and address challenges in biosignal processing, time series machine learning, and control engineering when designing closed-loop mobile health technologies. His work has been published in 20+ journal articles in venues including Brain Stimulation and the IEEE Transactions on Biomedical Engineering. He has been awarded several Best Paper, Poster, and Presentation awards at international conferences such as the IEEE-EMBS International Conference on Biomedical and Health Informatics. Through medical and industrial collaborations, his research is leading to clinical impact. In 2022, his findings formed the scientific basis for Food and Drug Administration Breakthrough Designation of non-invasive vagus nerve stimulation for patients with posttraumatic stress disorder.
