Back to search

Georgia Institute of Technology

Enabling Robust Online Processing of Physiological Signals Corrupted by External Vibrations

Abstract

dc:description.abstract

The objective of this research is to develop robust tools to process noninvasively measured physiological signals for hypovolemia estimation in out-of-hospital settings. Hypovolemia, or low blood volume, is a leading cause of preventable death in many out-of-hospital scenarios. A critical period to reduce hypovolemia-related deaths is during patient transport to medical facilities. With the timely nature of severe hypovolemia, accurate, real-time assessment of its progression can enable medical responders to promptly administer prehospital treatment. Previous research has shown that the seismocardiogram (SCG), which monitors cardiomechanical function noninvasively, can estimate hypovolemia progression more effectively than traditional vital signs. However, SCG signals are highly susceptible to vibration artifacts, which limits their use in prehospital settings where accurate and timely responses are essential. A key goal of this dissertation is to enable SCG processing in noisy environments through robust denoising and feature extraction tools that can operate in real-time or online settings. We first explored real-time SCG feature extraction, demonstrating that shared SCG signal dynamics can be leveraged to enhance real-time SCG feature tracking in noisy environments. We then investigated the effects of vehicular vibrations on SCG signals, designing a signal decomposition framework that used the SCG periodic structure to separate the two. Finally, we utilize generative diffusion models to learn the latent space of SCG beats and enhance SCG signals collected in ambulatory settings. The tools developed in this work aim to advance hypovolemia monitoring systems that utilize these physiological signals by making them suitable for noisy prehospital care and daily life settings.

Degree

thesis:*
Level thesis:degree_level
Doctoral
Department dc:contributor.department
Electrical and Computer Engineering
Grantor dc:publisher
Georgia Institute of Technology
Year dc:date.issued
2024

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Lin, David Jimmy
Advisor dc:contributor.advisor
  • Inan, Omer T.
Committee members dc:contributor.committeemember
  • Zhang, Ying
  • Anderson, David V.
  • Medda, Alessio
  • Hahn, Jin-Oh

Subjects

dc:subject × 11

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/76882

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Lin, David Jimmy. Enabling Robust Online Processing of Physiological Signals Corrupted by External Vibrations. Doctoral thesis, Georgia Institute of Technology, 2024. https://hdl.handle.net/1853/76882