Assessment of Frailty using a Wrist-worn Device

The COVID-19 pandemic has considerably shifted the focus of scientific research, speeding up the process of digitizing medical monitoring. Wearable technology is already widely used in medical research, as it has the potential to monitor the user's physical activity in daily life. Therefore, they are particularly appealing for evaluating older subjects in their environment to capture early signs of frailty and mobility-related problems. Early detection of abnormal physical performance and gait may help identify physically frail subjects at an increased risk of losing their independence but are still amenable to preventive interventions, such as structured exercise. This dissertation explores the use of body-worn accelerometers for automated assessment of frailty during walking activity. We proposed an automated process based on machine learning techniques, able to classify patients according to their frailty status using a set of gait-related parameters extracted from wearable sensors. Here, we highlight the importance of the position chosen for placing the sensors by comparing the performances of a wrist- against a lower back-worn sensor. Secondly, the dissertation analyzes the use of Continuous Wavelet Transform in combination with sensor-derived gait parameters for frailty status assessment. Continuous Wavelet Transform was applied to obtain time-frequency domain representations of gait signals. Here, the most statistically significant band-based information for frailty status assessment was identified by means of ANOVA and statistical t-test. Moreover, a Deep Convolutional Neural Network was trained and tested for identifying wavelet-based patterns to classify subjects as robust or non-robust, a category that includes both Fried’s frail and pre-frail phenotypes. Finally, the dissertation aims to explore in-home collected wearable-derived signals for frailty status assessment. Signal-derived traces were segmented. A subset of these segments was used to calculate the Subject Activity Level, an index to quantify how users were active throughout the day. The SAL index was then combined with gait-derived features to design a novel frailty status assessment algorithm.