A Novel Positive Transfer Learning Approach for Telemonitoring of Parkinson's Disease

Hyunsoo Yoon, Jing Li

Research output: Contribution to journalArticle

Abstract

Telemonitoring is the use of electronic devices to remotely monitor patients. Taking the Parkinson's disease (PD) as an example, the use of at-home testing device (AHTD) enables remote, internet-based measurement of PD vocal symptoms. Translating AHTD measurement into a unified PD rating scale (UPDRS) through predictive analytics enables cost-effective, convenient, and close tracking of PD progression. Building a predictive model between AHTD measurement and UPDRS is not straightforward because PD patients are highly heterogeneous, which requires patient-specific models. Learning a patient-specific model faces the challenge of limited data. Transfer learning (TL) tackles this challenge by leveraging other patients' information to make up the data shortage when modeling a target patient. Among different TL methods, the category of parameter transfer methods is more appropriate for the telemonitoring application because it transfers patient-specific model parameters but not patients' data. However, existing parameter transfer methods fall short because not every other patient's information is helpful and blind transfer causes the problem of negative transfer. To tackle this limitation, we propose a positive TL (PTL) method. We provide an in-depth theoretical study on the risk and condition for negative transfer to happen, which further drive the development of novel PTL algorithms that are robust to negative transfer. We apply PTL to predict UPDRS of 42 PD patients using their AHTD vocal measurement. PTL achieves significantly better accuracy compared with single learning and one-model-fits-all approaches.

Original languageEnglish (US)
JournalIEEE Transactions on Automation Science and Engineering
DOIs
StateAccepted/In press - Jan 1 2018

Fingerprint

Testing
Internet
Costs
Predictive analytics

Keywords

  • Data models
  • Diseases
  • Machine learning
  • Machine learning
  • Monitoring
  • negative transfer
  • Prediction algorithms
  • Predictive models
  • telemonitoring
  • transfer learning (TL).

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

@article{d176f8055a004d59a42bae723fc0c2cd,
title = "A Novel Positive Transfer Learning Approach for Telemonitoring of Parkinson's Disease",
abstract = "Telemonitoring is the use of electronic devices to remotely monitor patients. Taking the Parkinson's disease (PD) as an example, the use of at-home testing device (AHTD) enables remote, internet-based measurement of PD vocal symptoms. Translating AHTD measurement into a unified PD rating scale (UPDRS) through predictive analytics enables cost-effective, convenient, and close tracking of PD progression. Building a predictive model between AHTD measurement and UPDRS is not straightforward because PD patients are highly heterogeneous, which requires patient-specific models. Learning a patient-specific model faces the challenge of limited data. Transfer learning (TL) tackles this challenge by leveraging other patients' information to make up the data shortage when modeling a target patient. Among different TL methods, the category of parameter transfer methods is more appropriate for the telemonitoring application because it transfers patient-specific model parameters but not patients' data. However, existing parameter transfer methods fall short because not every other patient's information is helpful and blind transfer causes the problem of negative transfer. To tackle this limitation, we propose a positive TL (PTL) method. We provide an in-depth theoretical study on the risk and condition for negative transfer to happen, which further drive the development of novel PTL algorithms that are robust to negative transfer. We apply PTL to predict UPDRS of 42 PD patients using their AHTD vocal measurement. PTL achieves significantly better accuracy compared with single learning and one-model-fits-all approaches.",
keywords = "Data models, Diseases, Machine learning, Machine learning, Monitoring, negative transfer, Prediction algorithms, Predictive models, telemonitoring, transfer learning (TL).",
author = "Hyunsoo Yoon and Jing Li",
year = "2018",
month = "1",
day = "1",
doi = "10.1109/TASE.2018.2874233",
language = "English (US)",
journal = "IEEE Transactions on Automation Science and Engineering",
issn = "1545-5955",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - A Novel Positive Transfer Learning Approach for Telemonitoring of Parkinson's Disease

AU - Yoon, Hyunsoo

AU - Li, Jing

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Telemonitoring is the use of electronic devices to remotely monitor patients. Taking the Parkinson's disease (PD) as an example, the use of at-home testing device (AHTD) enables remote, internet-based measurement of PD vocal symptoms. Translating AHTD measurement into a unified PD rating scale (UPDRS) through predictive analytics enables cost-effective, convenient, and close tracking of PD progression. Building a predictive model between AHTD measurement and UPDRS is not straightforward because PD patients are highly heterogeneous, which requires patient-specific models. Learning a patient-specific model faces the challenge of limited data. Transfer learning (TL) tackles this challenge by leveraging other patients' information to make up the data shortage when modeling a target patient. Among different TL methods, the category of parameter transfer methods is more appropriate for the telemonitoring application because it transfers patient-specific model parameters but not patients' data. However, existing parameter transfer methods fall short because not every other patient's information is helpful and blind transfer causes the problem of negative transfer. To tackle this limitation, we propose a positive TL (PTL) method. We provide an in-depth theoretical study on the risk and condition for negative transfer to happen, which further drive the development of novel PTL algorithms that are robust to negative transfer. We apply PTL to predict UPDRS of 42 PD patients using their AHTD vocal measurement. PTL achieves significantly better accuracy compared with single learning and one-model-fits-all approaches.

AB - Telemonitoring is the use of electronic devices to remotely monitor patients. Taking the Parkinson's disease (PD) as an example, the use of at-home testing device (AHTD) enables remote, internet-based measurement of PD vocal symptoms. Translating AHTD measurement into a unified PD rating scale (UPDRS) through predictive analytics enables cost-effective, convenient, and close tracking of PD progression. Building a predictive model between AHTD measurement and UPDRS is not straightforward because PD patients are highly heterogeneous, which requires patient-specific models. Learning a patient-specific model faces the challenge of limited data. Transfer learning (TL) tackles this challenge by leveraging other patients' information to make up the data shortage when modeling a target patient. Among different TL methods, the category of parameter transfer methods is more appropriate for the telemonitoring application because it transfers patient-specific model parameters but not patients' data. However, existing parameter transfer methods fall short because not every other patient's information is helpful and blind transfer causes the problem of negative transfer. To tackle this limitation, we propose a positive TL (PTL) method. We provide an in-depth theoretical study on the risk and condition for negative transfer to happen, which further drive the development of novel PTL algorithms that are robust to negative transfer. We apply PTL to predict UPDRS of 42 PD patients using their AHTD vocal measurement. PTL achieves significantly better accuracy compared with single learning and one-model-fits-all approaches.

KW - Data models

KW - Diseases

KW - Machine learning

KW - Machine learning

KW - Monitoring

KW - negative transfer

KW - Prediction algorithms

KW - Predictive models

KW - telemonitoring

KW - transfer learning (TL).

UR - http://www.scopus.com/inward/record.url?scp=85056152326&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056152326&partnerID=8YFLogxK

U2 - 10.1109/TASE.2018.2874233

DO - 10.1109/TASE.2018.2874233

M3 - Article

JO - IEEE Transactions on Automation Science and Engineering

JF - IEEE Transactions on Automation Science and Engineering

SN - 1545-5955

ER -