Remote Medication Status Prediction for Individuals with Parkinson's Disease using Time-series Data from Smartphones

TL;DR

This paper introduces a Transformer-based approach to predict Parkinson's medication status from smartphone sensor data, achieving high accuracy and enabling personalized remote health monitoring.

Contribution

It presents a novel Transformer model that effectively predicts medication status from multi-modal smartphone data, addressing individual differences and out-of-lab data collection challenges.

Findings

Achieved high AUC scores for medication status prediction (0.95-0.98).

Demonstrated the effectiveness of patient-wise historical records and attention mechanisms.

Enabled personalized remote health sensing for Parkinson's disease.

Abstract

Medication for neurological diseases such as the Parkinson's disease usually happens remotely away from hospitals. Such out-of-lab environments pose challenges in collecting timely and accurate health status data. Individual differences in behavioral signals collected from wearable sensors also lead to difficulties in adopting current general machine learning analysis pipelines. To address these challenges, we present a method for predicting the medication status of Parkinson's disease patients using the public mPower dataset, which contains 62,182 remote multi-modal test records collected on smartphones from 487 patients. The proposed method shows promising results in predicting three medication statuses objectively: Before Medication (AUC=0.95), After Medication (AUC=0.958), and Another Time (AUC=0.976) by examining patient-wise historical records with the attention weights learned through a Transformer model. Our method provides an innovative way for personalized remote health sensing in a timely and objective fashion which could benefit a broad range of similar applications.