Masters Theses

Abstract

This study explores the integration of spectral mixtures of fetal heart rate (FHR) and uterine contraction (UC) signals to enhance the prediction of fetal acidosis, utilizing the CTU-CHB dataset. Several classification models were trained using two distinct oversampling techniques and inputs, demonstrating that models incorporating spectral mixtures significantly outperform those using raw signals. These models, particularly when combined with convolutional neural networks (CNNs) and attention mechanisms, achieved a notable F1-score of 0.98, with the highest model achieving an area under the Receiver Operating Characteristic (ROC) curve of 0.95. The research employs a variety of techniques including short-time Fourier transform and a butterworth filter to extract and smooth dominant frequency components from the signals. Additionally, an optimal pH threshold of 7.15 was identified for predicting fetal acidosis, striking a balance between sensitivity for acidotic and non-acidotic samples. The study also examines the impact of test split sizes on model performance, highlighting the importance of selecting appropriate size of training datasets to avoid overfitting and ensure robust evaluations. The findings indicate that the spectral combination of FHR and UC signals, processed through CNNs and enhanced by attention mechanisms, provides a robust method for predicting fetal acidosis. This methodological advancement promises significant improvements in prenatal care by enabling more accurate and reliable fetal health monitoring.

Advisor(s)

Corns, Steven

Committee Member(s)

Dagli, Cihan H., 1949-
Nicolosi, Gabriel

Department(s)

Engineering Management and Systems Engineering

Degree Name

M.S. in Engineering Management

Publisher

Missouri University of Science and Technology

Publication Date

Summer 2025

Journal article titles appearing in thesis/dissertation

Paper I, found on pages 30 to 46, has been submitted to 2025 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology.

Pagination

XII, 67 PAGES

Note about bibliography

Includes_bibliographical_references_(pages 65-66)

Rights

© 2025 Anusha Adhikari , All Rights Reserved

Document Type

Thesis - Open Access

File Type

text

Language

English

Thesis Number

T 12503

Download

Share

 
COinS