Mohamed S*, Ghaleb R, Mansour H and Hemeida A
Background: Data mining is making impressive breakthroughs in the field of medicine. Data mining can accomplish tasks such as classifying patients, finding associations between different features and exploring hidden patterns and trends in patient data that simplify and improve medical predictions. Optimization algorithms are used along with neural networks to mine datasets and classify them.
Methods: In this study, artificial neural networks were trained and optimized using various evolutionary optimization algorithms and then applied to classify 119 Egyptian patients with symptoms of coronary artery disease CAD.
Results: The mean age of the study population was 57.9 ± 11.162. The best optimization result was obtained by the PSOGSA algorithm, which correctly classified 97.2% of patients. This high classification rate also confirms that H. pylori IgG can be considered a factor of acute coronary syndrome (ACS).
Conclusion: Artificial neural networks can work as classifiers for patients with CAD. The high classification rates confirm that H. pylori infection is indeed a strong indicator for ACS. More approaches that use data mining in medicine should be investigated.
PDFShare this article
Cardiovascular Diseases & Diagnosis received 427 citations as per Google Scholar report
