ML in Medical Technology: Non-invasive Biosensors
Analysis of continuous sensor data for monitoring vital signs using Machine Learning.
Problem
Continuous monitoring of vital signs requires the conversion of raw sensor signals into precise clinical values.
Challenge
The signals are noisy, drift, and appear with a delay. Various factors such as sweat and individual physiology influence the values. The underlying biochemical processes are not yet fully understood.
Solution
We develop neural networks and symbolic regressors that learn this mapping.