Dina Katabi, Professor of Computer Science and Engineering
Director, Center for Wireless Networks and Mobile Computing,
MIT Department of Electrical Engineering and Computer Science
Prof Katabi’s talk started with a description of the global size of the problem of chronic diseases, the benefits of identifying acute exacerbations earlier plus the benefits of reducing hospitalisation. Furthermore she discussed the benefits of collecting increased volume and more robust data for clinical trials.
Prof Katabi showed the current state – wearing a pulse oximeter on your finger, wearing a BP cuff, wearing motion sensors, sleep lab tools. Not very sustainable or comfortable leading to adherence problems.
They have created a device called Emerald – a WiFi like box that does not need to be worn. It senses electromagnetic signals in the room and it uses advanced AI to monitor:
- gait, speed, falls
- ADLs + daily pattern
- vital signs
- heart rate
How can you detect these without sensors? We are surrounded by electromagnetic signals, and even by raising our hands we change those signals.
- sensed by altitude elevation
- Gait speed
- predictor of falls risk,
- important metric in parkinson’s, MS, frailty for disease monitoring
- predictor of exacerbation in CHF and COPD
- Sleep stages
- measuring reflectivity and then using an adversarial neural network to detect sleep stage
- Respiratory rate
- holding breath / apnoeic
- also see heart rate at peaks and troughs
Wearables become invisibles.