Abstract

This paper investigates the application of machine learning for cardiac disease prediction in resource-constrained healthcare settings. This study conducts an empirical study evaluating four classification algorithms (Support Vector Machine, Random Forest, Logistic Regression, Decision Tree) on a real-world dataset. The results demonstrate that SVM achieves the highest accuracy (91%) in identifying high-risk patients, highlighting its potential for clinical decision support. The study provides a detailed comparative analysis of model performance, discusses computational feasibility, and outlines practical deployment considerations. These findings contribute to the advancement of machine learning applications in African healthcare systems.

References

• I. Adewole et al. Disease pattern recognition with k-means clustering. Journal of Medical Informatics, 19(2):56–64, 2018.
• B. Afolabi et al. Predictive models using knn. Nigeria Health Information Systems Journal, 18(2):85–92, 2023.
• A. A. Akinyelu and A. O. Adewumi. Breast cancer prediction using machine learning techniques. Journal of Healthcare In- formatics, 15(3):123–132, 2020.
• Muhammad Ali, Youmin Zhou, and Lei Li. Machine learning models for asthma prediction in healthcare settings: accuracy, interpretability, and deployment strategies. Healthcare Infor- matics Research, 27(2):124–136, 2021.
• S. Ayodele et al. Expectation maximization in disease pre- diction. Nigerian Health Surveillance Reports, 13(1):11–20, 2023.
• L. Chen, T. Wang, and Y. Zhang. Comparative study of machine learning models for cardiovascular risk prediction: From logistic regression to ensemble learning. Artificial In- telligence in Medicine, 140:102478, 2023.
• A. Ekpenyong et al. Disease prediction using naive bayes. Nigerian Healthcare Journal, 16(1):34–43, 2023.
• R Ferdous and B Kane. Temporal causal discovery in chronic disease management in mali. Journal of Temporal Analysis, 9(2):56–66, 2023.
• Y. He et al. Contactless oxygen monitoring via radio waves in nigeria. Nigerian Medical Engineering Journal, 8(2):71–81, 2023.
• Y. Odeyemi et al. Gradient boosting machines in nigerian pa- tient data. Journal of Electronic Health Records, 19(4):56–63, 2022.
• A. Owolabi et al. Hypertension prediction using artificial neural networks. African Journal of Medical Informatics, 22(4):98–107, 2019.
• INSERM. Big data en sante´ des de´fis techniques et e´thiques a` relever. https://www.inserm.fr/dossier/big-data-en-sante/, 06 2022. http://images.google.com/.
• SM Jeong and DH Kim. Deep metric learning for ecg data in kenya. Journal of Cardiovascular Informatics, 12(3):105– 112, 2023.
• S. Kumar, R. Sharma, and P. Gupta. Deep learning-based car- diovascular disease prediction using convolutional neural net- works. Biomedical Signal Processing and Control, 59:101– 112, 2020.
• K Labach and O Kabiru. Dual event time transformer for ehr in rwanda. Journal of Health Data Science, 15(1):67–79, 2023.
• P. Mbatchou et al. Logistic regression in predicting hospital records. Journal of Medical Informatics, 25(2):99–106, 2021.
• E. Mensah et al. Ann for national health data analysis. Ghana National Health Insurance Scheme Reports, 20(3):205–213, 2023.
• F Ndiaye and M Diouf. K-means clustering for healthcare data analysis in senegal. African Journal of Health Informatics, 15(1):123–134, 2022.
• John Nguyen, Krish Patel, and Han Lee. Improving early diagnosis of alzheimer’s disease using machine learning: A focus on knn and decision tree models. Journal of Neurological Disorders, 35(1):45–58, 2022.
• D. Nkansah et al. Decision trees for patient outcome prediction. Ghana Health Journal, 17(1):66–75, 2022.
• J Nolin and A Sarr. Privacy-preserving ecg framework in senegal: The privecg model. Journal of Medical Data Privacy, 7(1):34–45, 2023.
• T. Obeng et al. Svm-based disease diagnosis system. Ghana Health Data Reports, 15(2):120–128, 2022.