Beyond Wearables: The Rise of Invisible AI-Powered Health Tech!

We face a “silver tsunami”: the number of older adults is increasing at an unprecedented rate, while shortages of both professional and family caregivers are growing. Can technology help? What if AI could analyze the wireless signals that bounce around a home and measure people’s physiological signals, diagnose new diseases, detect exacerbations of existing conditions, and track medication response, all without wearables and as patients go about their daily lives?

In this talk, I will describe how we can achieve this vision. Our work is enabled by two technologies. First, my lab has created a device resembling a Wi-Fi box that uses AI to analyze wireless reflections from individuals, enabling measurement of respiratory, cardiovascular, EEG, and musculoskeletal signals, without body contact. These devices are already used by pharmaceutical companies and deployed with patients living with Parkinson’s disease, Alzheimer’s disease, ALS, lupus, Crohn’s disease, and rheumatoid arthritis (RA), generating a large real-world dataset to train and validate the AI models. Second, I will explain how we adapted large language model (LLM) methods to interpret physiological data, establishing a foundation model for screening complex diseases and tracking medication response. I envision a future in which “the Invisibles”, a new class of AI-powered, contactless sensors, enable smart health environments and shift care from reactive to proactive, improving health and well-being.

Speaker: Dina Katabi, Massachussets Institute of Technology

Attend in person or online (See weblink)