A novel multi-task linear mixed model for smartphone-based telemonitoring

作者：

Highlights：

• Propose a new Multi-task Linear Mixed Model (MultiLMM) in a single framework.

• Develop an efficient and converging algorithm for parameter estimation.

• Provide theoretical analysis to demonstrate why MultiLMM outperforms LMM.

• Apply MultiLMM to telemonitoring data for patient prognosis.

摘要

•Propose a new Multi-task Linear Mixed Model (MultiLMM) in a single framework.•Develop an efficient and converging algorithm for parameter estimation.•Provide theoretical analysis to demonstrate why MultiLMM outperforms LMM.•Apply MultiLMM to telemonitoring data for patient prognosis.

论文关键词：Telemonitoring,Parkinson’s disease,Multi-task learning,Mixed effects model,mHealth,Smartphone

论文评审过程：Received 16 September 2019, Revised 29 July 2020, Accepted 29 July 2020, Available online 8 August 2020, Version of Record 8 September 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113809