Subgroup discovery of Parkinson's Disease by utilizing a multi-modal smart device system

TL;DR

This study demonstrates that combining multi-modal data from smart devices improves Parkinson's disease classification accuracy and enables the discovery of distinct PD subgroups.

Contribution

It provides a comprehensive evaluation of multi-modal assessments using smart devices for PD diagnosis and subgroup discovery, which was lacking in prior research.

Findings

Multi-modal data improves classification accuracy.

Distinct PD subgroups can be identified.

Multi-modal assessments outperform single-modal approaches.

Abstract

In recent years, sensors from smart consumer devices have shown great diagnostic potential in movement disorders. In this context, data modalities such as electronic questionnaires, hand movement and voice captures have successfully captured biomarkers and allowed discrimination between Parkinson's disease (PD) and healthy controls (HC) or differential diagnosis (DD). However, to the best of our knowledge, a comprehensive evaluation of assessments with a multi-modal smart device system has still been lacking. In a prospective study exploring PD, we used smartwatches and smartphones to collect multi-modal data from 504 participants, including PD patients, DD and HC. This study aims to assess the effect of multi-modal vs. single-modal data on PD vs. HC and PD vs. DD classification, as well as on PD group clustering for subgroup identification. We were able to show that by combining various modalities, classification accuracy improved and further PD clusters were discovered.