Medication for neurological diseases such as the Parkinson's disease usually
happens remotely at home, away from hospitals. Such out-of-lab environments
pose challenges in collecting timely and accurate health status data using the
limited professional care devices for health condition analysis, medication
adherence measurement and future dose or treatment planning. 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 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 status 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. We
believe 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.

本文提出了一种基于 Transformer 模型对帕金森病患者远程监测数据进行分析，以实现个性化远程医疗技术的方法，结果表明该方法在通过考察患者历史数据学习权重后在预测药物状态方面取得了良好的效果。