Design of Edge AI Wearables for a Personalized and Sustainable Healthcare in the IoT Era

Wearable devices are poised as the next frontier of innovation in the context of Internet-of-Things (IoT) to be able to provide personalized healthcare by interacting also with our everyday objects, which can be interconnected in ways that improve our lives and transform the medical industry. This new family of smart wearable devices provide a great opportunity for the integration of the next-generation of artificial intelligence (AI) based technologies in medical devices. However, major key challenges remain in achieving this potential due to inherent resource-constrained nature of wearable systems, coupled with their (in principle) limited computing power and data gathering requirements for Big Data medical applications, which can result in degraded and unreliable behavior and short lifetime. In this virtual seminar, Prof. Atienza will first discuss the challenges of ultra-low power (ULP) design and communication in smart IoT systems with particular focus, as key use case, on wearable devices for medical applications in the context of Big Data healthcare. Then, the opportunities for edge computing and edge AI in next-generation smart wearables will be highlighted as a scalable way to fully deliver the concept of personalized medicine. This new trend of smarter wearable architectures will need to combine new ULP multi-core embedded systems with neural network accelerators, as well as including energy-scalable software layers to monitor medical pathologies by event-driven monitoring. Overall, the next-generation of smart wearable devices in the healthcare context will be able to gracefully adapt the energy consumption and precision of the pathology detection outputs according to the requirements of our surrounding world and available energy at each moment in time, as living organisms do to operate efficiently in the real world.