Loading...
Development of a wearable fetal heart rate monitor: An analysis of fetal electrocardiogram extraction algorithms
Brake, Anna Banuelos, Xavier Simmons, Emma
Brake, Anna
Banuelos, Xavier
Simmons, Emma
Citations
Altmetric:
Files
Authors
Other Names
Location
Time Period
Advisors
Original Date
Digitization Date
Issue Date
2024-04-12
Type
Abstract
Genre
Keywords
Subjects (LCSH)
Citation
Brake, Anna; Banuelos, Xavier; Simmons, Emma. 2024. Development of a wearable fetal heart rate monitor: An analysis of fetal electrocardiogram extraction algorithms. -- In Proceedings: 23rd Annual Undergraduate Research and Creative Activity Forum. Wichita, KS: Wichita State University, p. 40.
Abstract
In a time of increased ease of access to medical care, there remain groups that lack sufficient access to healthcare. Maternal and fetal healthcare in rural and underserved areas remains insufficient. Inferior maternal and fetal healthcare outcomes can be correlated with decreased access to healthcare. Based on existing studies, congenital heart defects are the most common birth defect and cause of infant death and can be treated when found in a timely manner; this treatment can increase survival rates. Utilizing wearable technology allows greater access to maternal and fetal healthcare in rural and underserved communities, leading to an increase in positive medical outcomes. As the first step in development, extraction algorithms to separate fetal and maternal heart rate were explored. Four different extraction algorithms (BSS, TS, KF, and AFM) and various subtypes were analyzed to determine the most accurate separation technique to produce the clearest fetal heart rate. This study analyzed the FECGSYN Toolbox v1.0.0, an open-source realistic non-invasive foetal ECG (NI-FECG) generator. Different signal-to-noise ratios were utilized for the Muscular Artifact (6dB, 12dB, 18dB, 24dB) and Baseline Wander (-4dB, 0dB, 4dB, 8dB) data sets. Algorithms were analyzed according to the accuracy of the extracted fetal heart rate in comparison with the actual fetal heart rate. Fetal heart rate was successfully extracted from the synthesized data. For the BSS algorithm, extraction accuracy remained fairly consistent as signal-to-noise ratio increased, based on eight channel input. For the TS and KF algorithms, extraction accuracy generally increased as signal-to-noise ratio increased. For the AFM algorithm, extraction accuracy slightly increased as signal-to-noise ratio increased. Next steps of the project include determining the most accurate algorithm subtype that will remain accurate when exposed to different noise levels.
Table of Contents
Description
Presented to the 23rd Undergraduate Research and Creative Activity Forum (URCAF) held at the Rhatigan Student Center, Wichita State University, April 12, 2024.
Publisher
Wichita State University
Journal
Book Title
Series
URCAF;v.23