SmartEAR: Smartwatch-based Unsupervised Learning for Multi-modal Signal Analysis in Opportunistic Sensing Framework

TL;DR

This paper presents an unsupervised learning approach using clustering techniques on multi-modal physiological data from smartwatches to discover latent structures in naturalistic health monitoring settings.

Contribution

It introduces a novel framework leveraging k-means, GMM, and neural self-organizing maps for analyzing unlabeled multi-modal smartwatch data in opportunistic sensing.

Findings

Effective clustering of physiological signals into meaningful groups

Discovery of latent structures in unlabeled multi-modal data

Potential for improved health monitoring insights

Abstract

Wrist-bands such as smartwatches have become an unobtrusive interface for collecting physiological and contextual data from users. Smartwatches are being used for smart healthcare, telecare, and wellness monitoring. In this paper, we used data collected from the AnEAR framework leveraging smartwatches to gather and store physiological data from patients in naturalistic settings. This data included temperature, galvanic skin response (GSR), acceleration, and heart rate (HR). In particular, we focused on HR and acceleration, as these two modalities are often correlated. Since the data was unlabeled we relied on unsupervised learning for multi-modal signal analysis. We propose using k-means clustering, GMM clustering, and Self-Organizing maps based on Neural Networks for group the multi-modal data into homogeneous clusters. This strategy helped in discovering latent structures in our data.