Machine learning infused preventive healthcare for high-risk outpatient elderly

Publication Date

1-1-2018

Document Type

Conference Proceeding

Publication Title

Intelligent Systems and Applications: Proceedings of the 2018 Intelligent Systems Conference (IntelliSys) Volume 2

Editor

Kohei Arai, Supriya Kapoor, Rahul Bhatia

DOI

10.1007/978-3-030-01057-7_40

First Page

496

Last Page

511

Abstract

Medical treatment, lost work or productivity, and health care costs are significant burdens to the economy, families, and businesses. Preventive healthcare encourages health and averts disease or injuries by addressing factors that lead to the inception of a disease, and by detecting dormant conditions to reduce or cessation their progression. Preventive healthcare reduces the significant economic burden of disease in addition to improving the length and quality of outpatients’ lives. Machine Learning (ML) Infused Preventive Healthcare goes one step ahead by application of algorithms for collection of multi-scale clinical, biomedical, contextual, and environmental data about each outpatient (e.g., in Electronic Health Record (EHR)s, personal health records-PHR, etc.), unified and extensibility of metadata standards, and decision support tools to facilitate optimized patient-centered, evidence-based decisions. Through interweaving data, importantly, from traditional healthcare data sources such as outpatient Electronic Health Records (EHR) and revolutionary data sources such as mobile, voice and sensor generated outpatient contextual and lifestyle data, the machine learning (ML) infused preventive health care breeds new clinical pathways that are not only beneficial to the individual outpatients but can also improve overall population safety and health outcomes. In this research paper, we propose machine learning infused preventive health care and aims to solve one of the most important issues in the outpatient elderly healthcare “prevention of injuries and outpatient caring”. Finally, the paper presents a prototyping solution design as well as its application and certain experimental results.

Funding Sponsor

National Science Foundation

Keywords

Android, Association rule mining, Bagging, Boosting, Committee machine, Conditional random field, Decision tree, Electronic health records (HER), Health, Internet of Things (IoT), IOS, IoT architecture, Mobile sensors, Naïve Bayes classifier, Hidden Markov model, Outpatient, Preventive healthcare, Qualitative metrics, Real-time stream analytics, Sanjeevani electronic health records, Supervised machine learning, Apple Siri kit

Department

Computer Engineering

Link to Full Text

Share

COinS