Predictive Model of cardiovascular diseases based on Bayesian Networks
DOI:
https://doi.org/10.31637/epsir-2024-1804Keywords:
bayesian networks, cardiovascular disease, prevention, predictive models, BSEJ, KDB, naive Bayes, machine learningAbstract
Introduction: This study presents an analysis and comparison of several Bayesian network models for cardiovascular disease prediction, using clinical data to identify the most effective models. Methodology: The algorithms evaluated include Naive Bayes, TAN_cl, TAN_hcsp, FSSJ, BSEJ and KDB, which were trained and validated to measure their performance. A clinical dataset of patients, combining five public databases, was used to evaluate performance. The metrics used were accuracy, sensitivity, specificity, F1 Score and also a cross validation to ensure consistency of results. Results: The BSEJ model presented the best performance in all the metrics evaluated, standing out for its ability to eliminate irrelevant dependencies, thus maintaining an optimal balance between complexity and accuracy. Discussions: Bayesian networks offer a powerful tool for CVD prediction, as they not only provide accurate predictions, but also facilitate the interpretation of relationships between risk factors, which is crucial in the clinical setting. Conclusions: Bayesian networks, and in particular the BSEJ and KDB models, stand out for their effectiveness in predicting CVD, providing informed clinical decision support.
Downloads
References
Bravo, F. P., Del Barrio García, A. A., Gago Veiga, A. B., Gallego de la Sacristana, M. M., Pinero, M. R., Peral, A. G., Dzeroski, S. y Ayala, J. L. (2019). SMURF: Systematic Methodology for Unveiling Relevant Factors in Retrospective Data on Chronic Disease Treatments. IEEE Access, 7, 92598–92614. https://doi.org/10.1109/ACCESS.2019.2927429 DOI: https://doi.org/10.1109/ACCESS.2019.2927429
Bravo, F. P., García, A. A. D. B., Russo, L. M. S. y Ayala, J. L. (2020). SOFIA: Selection of Medical Features by Induced Alterations in Numeric Labels. Electronics, 9(9), 1492. https://doi.org/10.3390/electronics9091492 DOI: https://doi.org/10.3390/electronics9091492
Cardiovascular diseases (CVDs). (n.d.). Cardiovascular diseases. https://lc.cx/nFw6WA
Chicco, D. y Jurman, G. (2020). The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genomics, 21(1), 6. https://doi.org/10.1186/s12864-019-6413-7 DOI: https://doi.org/10.1186/s12864-019-6413-7
Ecuador acumula pacientes con enfermedades cardiovasculares | CEAP :: Centro de Estudios Asia-Pacífico (n.d.). https://lc.cx/x5lAR1
Enfermedades cardiovasculares - OPS/OMS | Organización Panamericana de la Salud. (n.d.). https://www.paho.org/es/temas/enfermedades-cardiovasculares
Heart Failure Prediction Dataset (n.d.). https://lc.cx/D2AHi-
Jahirul, M. (2010). Investigating the Performance of Naive- Bayes Classifiers and K- Nearest Neighbor Classifiers. International Conference on Convergence Information Technology (ICCIT 2007), 1541-1546. https://doi.org/10.1109/ICCIT.2007.148 DOI: https://doi.org/10.1109/ICCIT.2007.148
Kumar, S. y Sahoo, G. (2015). Classification of Heart Disease Using Naïve Bayes and Genetic Algorithm (pp. 269-282). https://doi.org/10.1007/978-81-322-2208-8_25 DOI: https://doi.org/10.1007/978-81-322-2208-8_25
La Carga de Enfermedades Cardiovasculares - OPS/OMS | Organización Panamericana de la Salud. (n.d.). https://www.paho.org/es/enlace/carga-enfermedades-cardiovasculares
Madden, M. G. (2009). On the classification performance of TAN and general Bayesian networks. Knowledge-Based Systems, 22(7), 489-495. https://doi.org/10.1016/j.knosys.2008.10.006 DOI: https://doi.org/10.1016/j.knosys.2008.10.006
Moscatelli, M., Manconi, A., Pessina, M., Fellegara, G., Rampoldi, S., Milanesi, L., Casasco, A. y Gnocchi, M. (2018). An infrastructure for precision medicine through analysis of big data. BMC Bioinformatics, 19(S10), 351. https://doi.org/10.1186/s12859-018-2300-5 DOI: https://doi.org/10.1186/s12859-018-2300-5
Nagavelli, U., Samanta, D. y Thomas, B. (2023). ML-Based Prediction Model for Cardiovascular Disease (pp. 91-98). https://doi.org/10.1007/978-981-19-4052-1_11 DOI: https://doi.org/10.1007/978-981-19-4052-1_11
Tovey, C. A. (1985). Hill Climbing with Multiple Local Optima. SIAM Journal on Algebraic Discrete Methods, 6(3), 384-393. https://doi.org/10.1137/0606040 DOI: https://doi.org/10.1137/0606040
Wang, X., Wang, F., Hu, J. y Sorrentino, R. (2014). Exploring joint disease risk prediction, 1180-1187.
Zhong, C., Pedrycz, W., Wang, D., Li, L. y Li, Z. (2016). Granular data imputation: A framework of Granular Computing. Applied Soft Computing, 46, 307-316. https://doi.org/10.1016/j.asoc.2016.05.006 DOI: https://doi.org/10.1016/j.asoc.2016.05.006
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Dayron Rumbaut Rangel, Milton Rafael Maridueña Arroyave
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who publish with this journal agree to the following terms:- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Non Commercial, No Derivatives Attribution 4.0. International (CC BY-NC-ND 4.0.), that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
