Processing and learning of cardiac signals: Application to heart failure
Author(s)
Aydemir, Varol Burak
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Heart Failure (HF) is a debilitating disorder contributing to nearly 300,000 deaths and
more than 800,000 hospitalizations each year in the US. The cost associated with HF expected to reach $70 billion by 2030. One of the driving factors of the mortality and cost
of HF is the high rate of readmission after patients’ initial hospitalization. Rehospitalizations can be reduced by remote sensor-based monitoring of HF patients. The motivation
for this work is the possibility of large-scale remote monitoring via non-invasive and unobtrusive devices that can measure cardiac function. Two cardiac signals acquired from
such devices are called ballistocardiography (BCG) and seismocardiography (SCG). BCG
and SCG measure the whole-body reaction forces and local chest vibrations in response
to cardiac ejection of the blood from the heart, respectively. The goal of this thesis is to
develop effective processing and learning methods for cardiac signals, such as BCG and
SCG, with an application to HF care. This work describes two processing and modeling
approaches for BCG and SCG signals for the task of classifying the HF patients’ clinical
status. In the first approach, a processing pipeline for BCG signals is developed to classify
clinical decompensation. Area under the receiver operating characteristic curve (AUC) of
0.78 is achieved in BCG data collected from HF patients. In the second effort, a processing
pipeline is developed for SCG signals to classify hemodynamic decompensation. AUC of
0.80 is achieved from SCG data collected in the hospital. The final piece of work aims
to handle the scarce labeled data in the cardiac signal domain. Recently, self-supervised
learning (SSL) methods saw great success with limited amount of labeled data. Motivated
by these successes, we propose a novel SSL task where a model is learned through matching the representations of simultaneously recorded cardiac signals. We show that the novel
SSL approach shows an improvement of 5% against the state-of-the-art SSL method in
scarce data settings. Furthermore, the approach has comparable performance in a variety
of settings with much lower data requirement (19 vs. 8800 subjects).
Sponsor
Date
2023-04-26
Extent
Resource Type
Text
Resource Subtype
Dissertation