Machine learning integration with robotics for biosensing : an automated medicine approach
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
Currently, the United States is experiencing a nursing shortage driven by the ageing population. This shortage is projected to worsen in the coming years as the percentage of retirement aged individuals increases. To mitigate the impacts of the nursing shortage, a system capable of automating certain daily tasks of nursing professionals is desired. To address this need, this dissertation focuses on utilizing the recent technological advancements in machine learning (ML) and robotics to design an automated biosignal monitoring robotic system. To accomplish this, electromyography (EMG) and photoplethysmography (PPG) are explored in detail to examine their capabilities in robotic control and vital sign monitoring. This dissertation consists of five chapters, each exploring different aspects of biosignal analysis and robotic integration. Chapter 2 explores EMG for simplified robotic control, using traditional ML models for classification. Chapter 3 extends this to medical EMG applications, employing deep network regression. Chapter 4 examines PPG signals for non-invasive blood pressure monitoring, utilizing traditional ML models for regression. Chapter 5 develops a PPG-integrated robotic system for autonomous vital sign monitoring. Finally, Chapter 6 summarizes the results, discusses their implications, and suggests a future research direction. This dissertation contributes to reducing the challenges associated with nursing shortages by providing an automated vital sign monitoring framework, potentially easing the burdens on nurses while enhancing patient care.
Table of Contents
DOI
PubMed ID
Degree
Ph. D.