Research Areas
AI for Community Health
A key challenge facing the US healthcare system is to improve health outcomes and reduce the cost of care for patients with chronic health conditions such as diabetes and heart failure. The increasing adoption of mobile health technology, in the form of smart watches, and of smart home technologies such as Amazon Alexa, create an unprecedented opportunity to address these challenges. These technologies provide a means to measure health-related variables from a patient’s daily life environment and deliver health-related messaging, motivation, and other forms of intervention to patients in real-time. Via this feedback loop, patients can be given new tools for achieving their health-related goals while working more effectively with care providers. Additionally, insights into the structural barriers to health improvements can be obtained to inform community development and public policy. In particular, these technologies can improve access to care in digital rural health, by allowing care providers to assay a patient's health status without inpatient visits or nursing outcall. Achieving these goals requires advances in multiple areas: machine learning methods that can leverage high volume streaming sensor data to infer states of health and behavior; novel adaptive intervention designs that can deliver just-in-time feedback; control systems modeling approaches that can characterize and optimize the closed-loop performance of such real-time interventions; and HCI and human factors research that can engage patients, clinicians, and stakeholders in designing health solutions via participatory design, while also addressing ethical and privacy concerns.
Mixed Reality Technologies for Diagnostics, Therapeutics and the Training of Healthcare Providers
The increasing capabilities of platforms for augmented and virtual reality, combined with their increasing adoption by users, has created the potential for new ways to collect health-related data, diagnose health conditions, and deliver novel digital therapeutics. Mixed reality headsets can incorporate a rich set of sensing modalities, including egocentric vision and eye tracking, with the potential to yield new insights into health-related behaviors and enable new forms of ecologically-grounded interventions. Moreover, the same platforms can be incorporated into educational contexts to facilitate personalized training and the development of novel simulation environments. In order to develop and exploit these capabilities requires advances in multiple areas: egocentric computer vision for quantifying the contextual and behavioral factors that lead to adverse health outcomes, VR/AR technologies for creating immersive and persuasive simulations, and generative AI methods to address the problem of scalable and effective content creation for virtual environments.