Kalp Krizi Riski Tahmininde Açıklanabilir Yapay Zeka Yaklaşımı

Yıl 2025, Cilt: 15 Sayı: 1, 1 - 15, 15.03.2025

Tülay Turan

https://doi.org/10.31466/kfbd.1473382

Öz

Bu çalışma, kalp krizi risklerinin analiz edilmesi ve doğru bir şekilde sınıflandırılması için açıklanabilir yapay zeka (XAI) tekniklerinin uygulanabilirliğini incelemeyi amaçlamaktadır. Kalp krizi risk faktörlerinin karmaşıklığı göz önünde bulundurulduğunda, geleneksel makine öğrenmesi modelleri genellikle klinik karar verme için gerekli olan şeffaflığı sağlamamaktadır. Bu araştırma, model tahminlerini açığa çıkarmak için özellikle SHAP (SHapley Additive exPlanations) gibi XAI tekniklerini dahil ederek bu boşluğu ele almaktadır. Çalışmada birden fazla veri tabanı taranarak 1319 hastanın 8 risk faktörüne ilişkin veriler elde edilmiştir. Kalp krizi sınıflandırması için altı farklı makine öğrenmesi algoritması kullanılarak tahmin modelleri geliştirilmiştir. Kalp krizi risk sınıflandırmasında XGBoost modeli %91,28 Accuracy, %90 Precision, %92 Recall ve %91 F1-Score ile en iyi tahmin değerlerini elde etmiştir. Ayrıca model algoritmaları AUC'a göre değerlendirildiğinde, XGBoost modelinin 0,91 doğruluk değeri ile en iyi sonucu elde edttiği görülmüştür. Random Forest özellik önem değerlendirmesinde değişkenler arasında tanıyı etkileyen en kritik değişkenin troponin olduğu görülmüştür. SHAP grafiklerinde de troponin (+4.19) en önemli risk faktörü olduğu görülmüştür. Bu araştırma, XAI'nın, karmaşık AI modelleri ile klinik uygulanabilirlik arasındaki boşluğu kapatma potansiyelini vurgulamakta ve gelecekteki çalışmaların AI destekli sağlık çözümlerini daha da rafine etmek ve doğrulamak için umut verici bir yönde ilerlemesini önermektedir.

Anahtar Kelimeler

Açıklanabilir Yapay Zeka, Kalp Krizi Risk Tahmini, Makine Öğrenmesi, XGBoost, SHAP

Explainable Artificial Intelligence Approach to Heart Attack Risk Prediction

Öz

This study examines the feasibility of explainable artificial intelligence (XAI) techniques for analyzing and accurately classifying heart attack risks. Given the complexity of heart attack risk factors, traditional machine learning models often do not provide the transparency needed for clinical decision-making. This research addresses this gap by incorporating XAI techniques, specifically SHAP (SHapley Additive exPlanations), to reveal model predictions. In this retrospective study, multiple databases were searched, and data on eight risk factors of 1319 patients were obtained. Prediction models have been developed using six different machine learning algorithms for heart attack classification. In heart attack risk classification, the XGBoost (eXtreme Gradient Boosting) model achieved the best predictive values with 91.28% Accuracy, 90% Precision, 92% Recall, and 91% F1-score. In addition, the model algorithms were evaluated according to AUC, and again, the XGBoost model achieved the best result 0.91. In the Random Forest Feature importance evaluation, troponin was the most critical variable affecting the diagnosis. SHAP graphs showed that troponin (+4.19) was the most critical risk factor. This research highlights the potential of XAI to bridge the gap between complex AI models and clinical applicability and suggests that future studies move in a promising direction to refine further and validate AI-powered healthcare solutions.

Anahtar Kelimeler

Explainable Artificial Intelligence, Heart Attack Risk Prediction, Machine Learning, XGBoost, SHAP

Kaynakça

• Abdulhussein A B, Bilgin T T. (2024). Comparison of Machine Learning Algorithms for Heart Disease Prediction. İstanbul Ticaret Üniversitesi Teknoloji ve Uygulamalı Bilimler Dergisi, 7(1), 133-146.
• Abubaker H, Muchtar F, Khairuddin A R, et al. (2024). Exploring Important Factors in Predicting Heart Disease Based on Ensemble-Extra Feature Selection Approach. Baghdad Science Journal, 21(2), 812-831.
• Aghamohammadi M, Madan M, Hong J K, Watson I. (2019). Predicting Heart Attack Through Explainable Artificial Intelligence. Computational Science, 11537.
• Ahsan M. (2022). Heart attack prediction using machine learning and XAI (Doctoral dissertation, Brac University).
• Akhiat Y, Manzali Y, Chahhou M, Zinedine A. (2021). A new noisy random forest-based method for feature selection. Cybernetics and Information Technologies, 21(2), 10-28.
• AlSagri H, Ykhlef M. (2020). Quantifying feature importance for detecting depression using random forest. International Journal of Advanced Computer Science and Applications, 11(5),628-635.
• Antwarg L, Miller R M, Shapira B, Rokach L. (2021). Explaining anomalies detected by autoencoders using Shapley Additive Explanations. Expert systems with applications, 186,115736.
• Arrieta A B, Díaz-Rodríguez N, Del Ser J, et al. (2020). Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information fusion, 58, 82-115.
• Ashraf M, Rizvi M A, Sharma H. (2019). Improved Heart Disease Prediction Using Deep Neural Network. Asian Journal of Computer Science and Technology, 8(2), 49–54.
• Chen T, Guestrin C. (2016, August). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery. San Francisco. California.
• Coolen A C C, Barrett J E, Paga P, Perez-Vicente C J. (2017). Replica analysis of overfitting in regression models for time-to-event data. Journal of physics A: mathematical and theoretical, 50(37), 375001.
• De Ville B. (2013). Decision trees. Wiley Interdisciplinary Reviews: Computational Statistics, 5(6), 448-455.
• Doğan Z, Küçükakçalı Z. (2023). Establishing a Model for the Classification of Heart Attack and Identification of Associated Risk Factors with Machine Learning Methods. ODÜ Tıp Dergisi, 10(3), 111-120.
• Dritsas E, Trigka M. (2024). Application of Deep Learning for Heart Attack Prediction with Explainable Artificial Intelligence. Computers, 13(10), 244.
• Ebashi S, Kodama A, Ebashı F. (1968). Troponin: 1. Preparation and physiological function. The Journal of Biochemistry, 64(4), 465-477.
• Ebashi S, Wakabayashi T, Ebashi F. (1971). Troponin and its components. The Journal of Biochemistry, 69(2), 441-445.
• Filatov V L, Katrukha A G, Bulargina T V, Gusev N B. (1999) Troponin: structure, properties, and mechanism of functioning. Biochemistry of Biokhimiia, 64, 969-985.
• Gunning D, Stefik M, Choi J, Miller T, Stumpf S, Yang G Z. (2019). XAI—Explainable artificial intelligence. Science robotics, 4(37), eaay7120.
• Harkulkar N, Nadkarni S, Patel B, Jadhav A. (2020). Heart Disease Prediction using CNN Deep Learning Model. Int. J. Res. Appl. Sci. Eng. Technol, 8, 875–881.
• Hasan M A M, Nasser M, Ahmad S, Molla K I. (2016). Feature selection for intrusion detection using random forest. Journal of information security, 7(3), 129-140.
• Hassija V, Chamola V, Mahapatra A, et al. (2024). Interpreting black-box models: a review on explainable artificial intelligence. Cognitive Computation, 16(1), 45-74.
• Hassija V, Chamola V, Mahapatra A, Singal A, Goel D, Huang K, Hussain A. (2024). Interpreting black-box models: a review on explainable artificial intelligence. Cognitive Computation, 16(1), 45-74.
• Hawkins D M. (2004). The problem of overfitting. Journal of chemical information and computer sciences, 44(1), 1-12.
• Hernandez J, Li Y, Rehg J M, Picard R W. (2019). BioWatch: A noninvasive wrist-based blood pressure monitor that incorporates training data from other subjects for machine learning. IEEE Journal of Biomedical and Health Informatics, 23(4), 1563-1570.
• Huang S, Cai N, Pacheco P P, Narrandes S, Wang Y, Xu W. (2018). Applications of support vector machine (SVM) learning in cancer genomics. Cancer genomics & proteomics, 15(1),41-51.
• Jakkula V. (2006). Tutorial on support vector machine (svm). School of EECS, Washington State University, 37(2.5), 3.
• Johnson K W, Torres Soto J, Glicksberg B S, et al. (2018). Artificial intelligence in cardiology. Journal of the American College of Cardiology, 73(23), 2935-2950.
• Katarya R, Meena S K. (2021). Machine learning techniques for heart disease prediction: a comparative study and analysis. Health and Technology, 11, 87-97.
• Kim Y, Kim Y. (2022). Explainable heat-related mortality with random forest and SHapley Additive exPlanations (SHAP) models. Sustainable Cities and Society, 79, 103677.
• Kırboğa K K, Küçüksille E U. (2023). Identifying Cardiovascular Disease Risk Factors in Adults with Explainable Artificial Intelligence. Anatolian Journal of Cardiology, 27(11), 657-663.
• Kumar K P, Thiruthuvanathan M M, Swathikiran K K, Chandra D R. (2024). Human AI: Explainable and responsible models in computer vision. Emotional AI and Human-AI Interactions in Social Network (pp. 237-254). Academic Press.
• Lee E T, Howard B V, Wang W, Welty, et al. (2006). Prediction of coronary heart disease in a population with high prevalence of diabetes and albuminuria: the Strong Heart Study. Circulation, 113(25), 2897-2905.
• Mangalathu S, Hwang S H, Jeon J S. (2020). Failure mode and effects analysis of RC members based on machine-learning-based SHapley Additive exPlanations (SHAP) approach. Engineering Structures, 219,110927.
• Marcus E, Teuwen J. (2024). Artificial intelligence and explanation: How, why, and when to explain black boxes. European Journal of Radiology, 173,111393.
• Mathews S M. (2019). Explainable Artificial Intelligence Applications in NLP, Biomedical, and Malware Classification: A Literature Review. CompCom 2019 Advances in Intelligent Systems and Computing.
• Moore A, Bell M. (2022). XGBoost, a novel explainable AI technique, in the prediction of myocardial infarction: a UK Biobank Cohort Study. Clinical Medicine Insights: Cardiology, 16, 11795468221133611.
• Movsessian A, Cava D G, Tcherniak D. (2022). Interpretable machine learning in damage detection using Shapley Additive Explanations. ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering, 8(2), 021101.
• Osman A I A, Ahmed A N, Chow M F, Huang Y F, El-Shafie A. (2021). Extreme gradient boosting (Xgboost) model to predict the groundwater levels in Selangor Malaysia. Ain Shams Engineering Journal, 12(2), 1545-1556.
• Parsa A B, Movahedi A, Taghipour H, Derrible S, Mohammadian A K. (2020). Toward safer highways, application of XGBoost and SHAP for real-time accident detection and feature analysis. Accident Analysis Prevention, 136: 105405.
• Pitchal P, Ponnusamy S, Soundararajan V. (2024). Heart disease prediction: Improved quantum convolutional neural network and enhanced features. Expert Systems with Applications, 249, 123534.
• Pothuganti S. (2018). Review on over-fitting and under-fitting problems in Machine Learning and solutions. International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, 7(9), 3692-3695.
• Qiu Y, Zhou J, Khandelwal M, Yang H, Yang P, Li, C. (2022). Performance evaluation of hybrid WOA-XGBoost, GWO-XGBoost and BO-XGBoost models to predict blast-induced ground vibration. Engineering with Computers, 38(5), 4145-4162.
• Rigatti S J. (2017). Random forest. Journal of Insurance Medicine, 47(1), 31-39.
• Sharma S, Parmar M. (2020). Heart diseases prediction using deep learning neural network model. Int. J. Innov. Technol. Explor. Eng. (IJITEE), 9, 2244–2248.
• Takci H. (2022). Performance-enhanced KNN algorithm-based heart disease prediction with the help of optimum parameters. Journal of the Faculty of Engineering and Architecture of Gazi University, 38(1).
• Thiele H, Rach J, Klein N, et al. (2021). Optimal timing of invasive angiography in stable non-ST-elevation myocardial infarction: the Leipzig Immediate versus early and late PercutaneouS coronary Intervention triAl in NSTEMI (LIPSIA-NSTEMI Trial). European heart journal, 33(16), 2035-2043.
• URL-1: https://www.who.int/health-topics/cardiovascular-diseases#tab=tab_1 (Date Accessed: 20 November 2023).
• URL-2:https://www.kaggle.com/datasets/bharath011/heart-disease-classification-dataset/data?select=Heart+Attack.csv (Date Accessed: 11 December 2023).
• Vatansever B, Aydın H, Çetinkaya A. (2021). Heart Disease Prediction with Machine Learning Algorithm Using Feature Selection by Genetic Algorithm. Bilim, Teknoloji ve Mühendislik Araştırmaları Dergisi, 2(2), 67-80.
• Wang H, Yang F, Luo, Z. (2016). An experimental study of the intrinsic stability of random forest variable importance measures. BMC bioinformatics, 17, 1-18.
• Zhang M L, Zhou Z H. (2007). ML-KNN: A lazy learning approach to multi-label learning. Pattern recognition, 40(7), 2038-2048.