THE IMPORTANCE OF EXPLAINABILITY IN MEDICAL AI AND PREDICTOR REDUCTION IN CARDIOVASCULAR RISK ASSESSMENT

Arslonbek Saidov; Daler Sharipov

Authors

Saidov Arslonbek Davron o'g'li Tashkent Information Technologies University
Sharipov Daler Kuchkorovich Tashkent Information Technologies University

Keywords:

Explainable AI (XAI), Medical AI, Predictor Reduction, Cardiovascular Risk Assessment, Myocardial Infarction, Adverse Cardiovascular Events, SHAP (Shapley Additive Explanations), LIME (Local Interpretable Model-agnostic Explanations)

Abstract

This article discusses the critical importance of explainability in medical AI, particularly in the context of cardiovascular risk prediction. It highlights the necessity of transparent AI systems for clinician trust, patient-centered care, regulatory compliance, and bias detection. The role of predictor reduction using XAI methodologies, such as SHAP and LIME, is emphasized as a means to enhance model efficiency and interpretability. The article also explores the challenges of achieving explainability, the evolving landscape of XAI research, and the future prospects of integrating explainable models with real-world healthcare systems. Ultimately, it underscores the need for interdisciplinary collaboration to ensure AI-driven medical decision-making is both effective and ethically sound.

References

Skouteli C, Prenzas N, Kakas A, Pattichis CS. Explainable AI Modeling in the Prediction of Cardiovascular Disease Risk. Stud Health Technol Inform. 2024 Aug 22;316:978-982. doi: 10.3233/SHTI240574. PMID: 39176955.

Gupta, J., Seeja, K.R. A Comparative Study and Systematic Analysis of XAI Models and their Applications in Healthcare. Arch Computat Methods Eng 31, 3977–4002 (2024). https://doi.org/10.1007/s11831-024-10103-9

Chaddad A, Peng J, Xu J, Bouridane A. Survey of Explainable AI Techniques in Healthcare. Sensors (Basel). 2023 Jan 5;23(2):634. doi: 10.3390/s23020634. PMID: 36679430; PMCID: PMC9862413.

Talaat FM, Elnaggar AR, Shaban WM, Shehata M, Elhosseini M. CardioRiskNet: A Hybrid AI-Based Model for Explainable Risk Prediction and Prognosis in Cardiovascular Disease. Bioengineering (Basel). 2024 Aug 12;11(8):822. doi: 10.3390/bioengineering11080822. PMID: 39199780; PMCID: PMC11351968.

J. Duell, X. Fan, B. Burnett, G. Aarts and S. -M. Zhou, "A Comparison of Explanations Given by Explainable Artificial Intelligence Methods on Analysing Electronic Health Records," 2021 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI), Athens, Greece, 2021, pp. 1-4, doi: 10.1109/BHI50953.2021.9508618. keywords: {Machine learning algorithms;Conferences;Biological system modeling;Decision making;Medical services;Machine learning;Predictive models;Explainable AI;Black-box;Glass-box;Machine Learning;Electronic Health Records},

Hempel, P., Ribeiro, A.H., Vollmer, M. et al. Explainable AI associates ECG aging effects with increased cardiovascular risk in a longitudinal population study. npj Digit. Med. 8, 25 (2025). https://doi.org/10.1038/s41746-024-01428-7

Ahmed Salih, PhD ,https://orcid.org/0000-0002-0871-8282 [email protected],

Ilaria Boscolo Galazzo, PhD, Polyxeni Gkontra, PhD https://orcid.org/0000-0001-8828-6143, Aaron Mark Lee, PhD https://orcid.org/0000-0001-6008-5366, Karim Lekadir, PhD, Zahra Raisi-Estabragh, PhD https://orcid.org/0000-0002-7757-5465, Steffen E. Petersen, DPhil https://orcid.org/0000-0003-4622-5160Author Info & Affiliations Circulation: Cardiovascular Imaging Volume 16, Number 4,https://doi.org/10.1161/CIRCIMAGING.122.014519

Salih, A.M., Galazzo, I.B., Gkontra, P. et al. A review of evaluation approaches for explainable AI with applications in cardiology. Artif Intell Rev 57, 240 (2024). https://doi.org/10.1007/s10462-024-10852-w

Sheu RK, Pardeshi MS. A Survey on Medical Explainable AI (XAI): Recent Progress, Explainability Approach, Human Interaction and Scoring System. Sensors (Basel). 2022 Oct 21;22(20):8068. doi: 10.3390/s22208068. PMID: 36298417; PMCID: PMC9609212.

Guleria, P.; Naga Srinivasu, P.; Ahmed, S.; Almusallam, N.; Alarfaj, F.K. XAI Framework for Cardiovascular Disease Prediction Using Classification Techniques. Electronics 2022, 11, 4086. https://doi.org/10.3390/electronics11244086