Detecting Parkinson's Disease from interactions with a search engine: Is expert knowledge sufficient?

TL;DR

This study demonstrates that mouse tracking data during search engine interactions can effectively distinguish Parkinson's disease patients from non-diseased users, with combined expert and unsupervised features achieving high diagnostic accuracy.

Contribution

The paper introduces a novel methodology for PD detection using mouse tracking data and compares expert-designed features with unsupervised learned features for improved classification.

Findings

Achieved an AUC of 0.92 using combined features.

Expert and auto-generated features both contribute to detection performance.

Mouse tracking can help identify early-stage Parkinson's disease.

Abstract

Parkinson's disease (PD) is a slowly progressing neurodegenerative disease with early manifestation of motor signs. Recently, there has been a growing interest in developing automatic tools that can assess motor function in PD patients. Here we show that mouse tracking data collected during people's interaction with a search engine can be used to distinguish PD patients from similar, non-diseased users and present a methodology developed for the diagnosis of PD from these data. A main challenge we address is the extraction of informative features from raw mouse tracking data. We do so in two complementary ways: First, we manually construct expert-recommended informative features, aiming to identify abnormalities in motor behaviors. Second, we use an unsupervised representation learning technique to map these raw data to high-level features. Using all the extracted features, a Random Forest classifier is then used to distinguish PD patients from controls, achieving an AUC of 0.92, while results using only expert-generated or auto-generated features are 0.87 and 0.83, respectively. Our results indicate that mouse tracking data can help in detecting users at early stages of the disease, and that both expert-generated features and unsupervised techniques for feature generation are required to achieve the best possible performance