Machine learning and control methodologies with applications to medical computing
Abstract/Contents
- Abstract
- Machine Learning (ML) in adult healthcare has demonstrated significant benefits in a wide range of applications, and provides a proof-of-concept for extended work in pediatrics. In contrast, ML in pediatric healthcare is a relatively immature field with huge potential for improving quality of patient care. Tools built using a combination of unique data sources, with novel theoretical and practical approaches, provide significant benefits when delivered to the hospital via the Electronic Medical Record (EMR). Statistical analysis of detailed in-hospital datasets enable standardizing patient care and construct a sound foundation for data-driven and automated decision support systems. Similarly, machine learning methodologies enable predictive models to optimize patient outcomes, and create frameworks for analyzing variable importances to improve efficiency in resource-limited critical care environments. Beyond the EMR, continuous bedside monitors record physiological waveforms to track each patient's state throughout their hospitalization. These dense, real-time data enable continuous provision of results, alerts, and predictions, but require novel deep learning models and adaptations to process and interpret. Convolutional Neural Networks (CNNs) provide a structured framework for processing such data sources, which is powerful yet flexible enough to adapt to a wide range of applications. Both in and outside the hospital environment, wearable devices provide similar physiological data streams to the in-hospital monitors. Such devices and the Wireless Body-Area Networks (WBANs) they comprise support remote patient monitoring in the ``e-Health'' paradigm. These systems are supported and enabled by theoretical and practical developments in fundamental wireless communication and queuing theory, yet require particular considerations for application to patient monitoring and care. In this thesis, I present novel contributions to both theoretical models and practical applications for each of these three lines of research. Each of these either directly addresses or indirectly supports a specific use case in the pediatric hospital environment.
Description
Type of resource | text |
---|---|
Form | electronic resource; remote; computer; online resource |
Extent | 1 online resource. |
Place | California |
Place | [Stanford, California] |
Publisher | [Stanford University] |
Copyright date | 2019; ©2019 |
Publication date | 2019; 2019 |
Issuance | monographic |
Language | English |
Creators/Contributors
Author | Miller, Daniel Roy |
---|---|
Degree supervisor | Bambos, Nicholas |
Thesis advisor | Bambos, Nicholas |
Thesis advisor | Ashlagi, Itai |
Thesis advisor | Poon, Ada Shuk Yan |
Degree committee member | Ashlagi, Itai |
Degree committee member | Poon, Ada Shuk Yan |
Associated with | Stanford University, Department of Electrical Engineering. |
Subjects
Genre | Theses |
---|---|
Genre | Text |
Bibliographic information
Statement of responsibility | Daniel Roy Miller. |
---|---|
Note | Submitted to the Department of Electrical Engineering. |
Thesis | Thesis Ph.D. Stanford University 2019. |
Location | electronic resource |
Access conditions
- Copyright
- © 2019 by Daniel Roy Miller
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...