Session

Aging Independently Using Machine Learning, Advanced Sensors and Algorithms

Kevin Shaw · Algorithmic Intuition

The population of Aging Adults is growing rapidly throughout the world, in the US alone more than 11,000 adults turn 65 each day. Staying at home as one ages is unquestionably cheaper and more healthful, but it brings risks to those living alone or needing care.

Homecare is usually provided by caregivers visiting on a regular basis to check and provide care, in most cases though they only visit a couple hours a week. Using technology we can fill in the gaps, keep aging adults safer and providing both them and their families greater peace-of-mind.

Now with Machine / Deep Learning approaches not only can richer measurements of vital statistics be made, but also daily activities and patterns of wellness can be tracked. Walking quality and its decline can be monitored and in some cases falls anticipated.

Activities of Daily Living (ADLs) can be identified and trends flagged, allowing caregivers (either provider or family) to have honest conversations about how care should to change. Illnesses can be flagged earlier and eating declines can be remedied pro-actively, leading to longer lives.

Critically, this brings direct health information to the consumer both immediately and regularly. No longer are they left to guess if they are doing well or to wait two months for their next few minutes with a doctor. They can monitor their own health and make their own decisions relative to population-based and personal histories.

All of this is only possible by fusing sensor information together and using the strong pattern matching capabilities of Machine/Deep Learning techniques. In this talk, we will discuss advances in in-home elder monitoring and the way advanced sensors and algorithms are driving the change for better care.