PREDICTING CARDIOVASCULAR DISEASES BY COMBINING EVIDENCES USING DEMPSTER SHAFER THEORY

Thái Hà Dương Nguyễn1,, Đình Khiết Lê1, Trần Đạt Lê1, Thị Thu Phương Phạm1, Thị Huế Ngô1, Thị Ngọc Lan Phan1, Thanh Xuân Phạm1
1 VNU Hanoi-University of Medicine and Pharmacy

Main Article Content

Abstract

Cardiovascular diseases (CVDs) are the leading cause of death worldwide, accounting for 31% of all deaths. The early diagnosis and stage of the diseases greatly support the treatment process, limiting the evolutions, complications and deadth rate. This process through the analysis of information, evidence, clinical examination symptoms, subclinical by experts, medical doctors. Recently, to contribute to the diagnostic process, artificial intelligence has been applied to speed up the analysis and processing process. These methods mostly use probability theory with the central role being Bayes' theorem. In this study, we also predicted cardiovascular diseases with data science approach, but followed another way – evidence-based integration using Dempster Shafer theory. In particular, each symptom is considered a evidence about the disease with some degree of uncertainty. Dempster combine is used to synthesize the evidence. The degree of uncertainty of each piece of evidence will be optimized by the gradient descent optimization algorithm. Preliminary results show that this new method not only has a significant improvement in predictability when compared with Bayesian but also shows the certainty of each symptom in the diagnostic process. These results allow expectations for the clinical support of the method as well as the potential application of data science to the field of medicine.

Article Details

References

1. Virani, Salim S., et al. "Heart disease and stroke statistics—2021 update: a report from the American Heart Association." Circulation 143.8 (2021): e254-e743.
2. Centers for Disease Control and Prevention. "Heart Disease Facts" (2022).
3. Chouard, T. (2016). The Go Files: AI computer wraps up 4-1 victory against human champion. Nature News.
4. Sorkin, R. D., & Woods, D. D. (1985). Systems with human monitors: A signal detection analysis. Human-computer interaction, 1(1), 49-75.
5. Fatima, M., & Pasha, M. (2017). Survey of machine learning algorithms for disease diagnostic. Journal of Intelligent Learning Systems and Applications, 9(01), 1.
6. Jackins, V., Vimal, S., Kaliappan, M. et al. AI-based smart prediction of clinical disease using random forest classifier and Naive Bayes. J Supercomput 77, 5198–5219 (2021)
7. Peñafiel, Sergio, et al. "Applying Dempster–Shafer theory for developing a flexible, accurate and interpretable classifier." Expert Systems with Applications 148 (2020): 113262.
8. Ruder, Sebastian. "An overview of gradient descent optimization algorithms." arXiv preprint arXiv:1609.04747 (2016).