Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form
D-Scholarship @ Pitt Banner and Links to Homepage

Towards Unveiling the Potential of the 12-Lead Electrocardiogram in Predicting Acute Coronary Syndrome via Machine Learning

Bouzid, Zeineb (2024) Towards Unveiling the Potential of the 12-Lead Electrocardiogram in Predicting Acute Coronary Syndrome via Machine Learning. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

[img] PDF
Restricted to University of Pittsburgh users only until 3 June 2026.
Download (21MB) | Request a Copy

Abstract

Nearly 7 million Americans visit the emergency department annually with a chief complaint of chest pain. Approximately 10\% of those patients have an acute disruption in blood supply to the heart attributed to underlying atherosclerotic disease in the coronary arteries, a life-threatening condition referred to as acute coronary syndrome. The prompt identification of acute coronary syndrome is a key challenge in clinical practice. The 12-lead electrocardiogram is readily available during initial patient evaluation, but current rule-based interpretation approaches lack sufficient accuracy. In this research, we utilize advanced signal processing techniques and machine learning methods to study the prognostic value of the electrocardiogram in early screening for acute coronary syndrome and related adverse outcomes. We investigate the use of statistical, deep learning and hybrid techniques to (1) detect and localize artery occlusions, (2) improve non-invasive risk stratification in chest pain patients, and (3) enhance the sensitivity and precision of occlusion myocardial infarction (a particularly deadly subcategory of acute coronary syndrome) identification. These projects jointly aim at developing an improved interpretable decision-support system for electrocardiograms to alert clinicians in real-time to acute coronary syndrome and bypass suboptimal time-consuming biomarker-driven tests. This would increase the available therapeutic window for initiating adequate therapy in distressed patients and reduce unnecessary prolonged surveillance in non-specific chest pain. Such advances would lower costs associated with admission and more involved tests, and produce better outcomes for patients.

Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Bouzid, Zeinebzeb12@pitt.eduzeb120000-0002-1095-4444
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairSejdic, Ervinesejdic@pitt.edu
Committee MemberAkcakaya, Muratakcakaya@pitt.edu
Committee MemberDallal, Ahmedahd12@pitt.edu
Committee MemberZhan, Liangliang.zhan@pitt.edu
Committee MemberAl-Zaiti, Salahssa33@pitt.edu
Date: 3 June 2024
Date Type: Publication
Defense Date: 27 March 2024
Approval Date: 3 June 2024
Submission Date: 25 March 2024
Access Restriction: 2 year -- Restrict access to University of Pittsburgh for a period of 2 years.
Number of Pages: 162
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Electrical and Computer Engineering
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: electrocardiogram, acute coronary syndrome, occlusion myocardial infarction, mortality, risk stratification, machine learning, deep learning
Date Deposited: 03 Jun 2024 14:39
Last Modified: 03 Jun 2024 14:39
URI: http://d-scholarship.pitt.edu/id/eprint/45900

Metrics

Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item