Tools
Tools
Adolfo, Cid Mathew (2024) Sample Reduction for Physiological Data Using Principal Component Analysis in Artificial Neural Network. PhD thesis, University of Gloucestershire. doi:10.46289/8MM26HP8
|
Text (Final Copy)
14687 ADOLFO (2024) PhD_Sample Reduction for Physiological Data Using Principal Component Analysis in Artificial Neural Network..pdf - Accepted Version Available under License All Rights Reserved. Download (42MB) | Preview |
Abstract
In the realm of biomedical applications, the analysis of big data holds immense promise, particularly in aiding medical practitioners with crucial interpretations. However, ensuring the integrity of multi-dimensional datasets and achieving high classification accuracy are paramount challenges. The quality of data sources presents a significant threat, with issues like abnormal, random and biased sampling posing obstacles, especially in machine learning contexts. Addressing these challenges is essential, particularly in biomedical applications reliant on accurate classification and prediction, such as physiological signal analysis utilizing Artificial Neural Networks (ANNs). This study proposes a novel approach, utilizing Principal Component Analysis–Sample Reduction Process (PCA-SRP), to preprocess datasets and enhance ANN model accuracy. We discuss the theoretical underpinnings of this methodology, followed by empirical validation using publicly available physiological and L/R Motor Movement (MM)EEG datasets. The analysis demonstrates the efficacy of PCA-SRP in cleansing datasets and improving ANN classification performance, with significant enhancements observed across various performance metrics. Notably, our approach achieves up to a 7% increase in accuracy in classifying L/R motor movement EEG signals, as validated through Python implementation.
| Item Type: | Thesis (PhD) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Thesis Advisors: |
|
||||||||||||
| Uncontrolled Keywords: | biomedical applications; analysis; medical practitioners; data sources; physiological signal analysis; Artificial Neural Networks (ANNs); Principal Component Analysis–Sample Reduction Process (PCA-SRP), | ||||||||||||
| Subjects: | Q Science > QA Mathematics > QA76 Computer software R Medicine > RA Public aspects of medicine |
||||||||||||
| Divisions: | Schools and Research Institutes > School of Business, Computing and Social Sciences | ||||||||||||
| Depositing User: | Anne Pengelly | ||||||||||||
| Date Deposited: | 14 Jan 2025 12:27 | ||||||||||||
| Last Modified: | 14 Jan 2025 12:47 | ||||||||||||
| URI: | https://eprints.glos.ac.uk/id/eprint/14687 |
University Staff: Request a correction | Repository Editors: Update this record
