Ensemble Regression and Explainable AI for Predicting Resource Utilization Efficiency in 6G-Enabled Smart Healthcare Systems

Abstract

Optimal management of resources will be founda- tional to future health systems designing around 6G technology. The combination of ultra-reliable, very low-latency connectivity with autonomous AI-based decision-making will automate many of the operational aspects of healthcare delivery. The research presented here establishes a total machine learning framework that can implement various regression analysis techniques and ensemble models to predict Resource Utilization Efficiency (RUE). We constructed a diverse dataset from clinical, operational, and telecommunications-based variables and utilized multiple data preprocessing techniques (imputation, encoding, scaling, and outlier correcting) to optimize the training of our six benchmarked regression analyses: Linear Regression, Random Forest, Gradient Boost, XGBoost, Support Vector Regression, and K-Nearest Neighbors. Results demonstrated that tree-based models achieved the highest predictive accuracy, with Random Forest, Gradient Boosting, and XGBoost consistently outperforming linear and kernel-based approaches. To further enhance performance, ensemble learning methods (averaging, blending, and stacking) were employed, with stacking ensembles delivering the best overall results (MSE = 1.86 × 10−5, R2 = 0.9998). To produce robust models through hyperparameter tuning with GridSearchCV and Bayesian optimization; the SHAP analysis method was conducted to provide interpretation to the decision process, revealing that Network Performance (speed), Length Of Stay and Health Status were the most significant variables in predicting RUE. Promoting predictiveness while maintaining transparency provides a concrete, interpretable decision support tool for healthcare decision-makers. With the proposed framework, intelligent; sustainable and explainable; 6G supported Healthcare Management may continue to expand to include federated learning; real-time implementation; and multi-modal data.

Keywords

• 6G healthcare
• ensemble regression
• explainable AI
• machine learning
• resource utilization efficiency
• smart medical systems

6G Etkin Akıllı Sağlık Sistemlerinde Kaynak Kullanım Verimliliğini Tahmin Etmek İçin Topluluk Regresyonu ve Açıklanabilir Yapay Zeka

Abstract

Verimli kaynak yönetimi, özellikle ultra güvenilir, düşük gecikmeli bağlantı ve yapay zeka odaklı karar almanın klinik operasyonları dönüştürmesi beklenen 6G etkin ortamlarda, yeni nesil sağlık sistemlerinin temel taşıdır. Bu çalışma, gelişmiş regresyon modelleri, topluluk stratejileri ve açıklanabilir yapay zeka tekniklerini entegre ederek Kaynak Kullanım Verimliliğini (RUE) tahmin etmek için kapsamlı bir makine öğrenimi çerçevesi sunmaktadır. Klinik, operasyonel ve ağ ile ilgili değişkenlerden oluşan çeşitli bir veri kümesi, sağlam model eğitimi sağlamak için yükleme, kodlama, ölçekleme ve aykırı değer işleme yoluyla ön işleme tabi tutulmuştur. Altı regresyon algoritması; Doğrusal Regresyon, Rastgele Orman, Gradient Boosting, XGBoost, Destek Vektörü Regresyonu ve K-En Yakın Komşular sistematik olarak kıyaslanmıştır. Sonuçlar, ağaç tabanlı modellerin en yüksek tahmin doğruluğunu elde ettiğini; Rastgele Orman, Gradient Boosting ve XGBoost'un doğrusal ve çekirdek tabanlı yaklaşımlardan sürekli olarak daha iyi performans gösterdiğini göstermiştir. Performansı daha da artırmak için, en iyi genel sonuçları (MSE = 1,86x10^-5, R^2 = 0,9998) veren yığın öğrenme yöntemleri (ortalama alma, karıştırma ve istifleme) kullanıldı. GridSearchCV ve Bayes optimizasyonu aracılığıyla hiperparametre ayarı, model sağlamlığını iyileştirdi. Yorumlanabilirliği sağlamak için SHAP analizi uygulandı ve Ağ Hızı, Kalış Süresi ve Sağlık Durumunun RUE tahminlerini yönlendiren en etkili faktörler olduğunu ortaya koydu. Tahmini performansı şeffaflıkla birleştirerek, bu çalışma sağlık yöneticileri için güvenilir ve yorumlanabilir bir karar destek aracı sağlamaktadır. Önerilen çerçeve, federasyonlu öğrenme, gerçek zamanlı dağıtım ve çok modlu veri entegrasyonu gibi potansiyel uzantılarla akıllı, sürdürülebilir ve açıklanabilir 6G özellikli sağlık yönetimine giden yolu açmaktadır.

Keywords

• 6G sağlık hizmeti
• topluluk regresyonu
• açıklanabilir yapay zeka
• makine öğrenimi
• kaynak kullanım verimliliği
• akıllı tıbbi sistemler

References

1. [1] Khan, I. A., Salam, A., Ullah, F., Amin, F., Tabrez, S., Faisal, S., & Choi, G. S. (2024). Big data analytics model using artificial intelligence (AI) and 6G technologies for healthcare. IEEE Access, 12, 97924-97937.
2. [2] Sharma, N., & Kaushik, P. (2025). Integration of AI in healthcare systems—A discussion of the challenges and opportunities of integrating AI in healthcare systems for disease detection and diagnosis. AI in Disease detection: advancements and applications, 239-263.
3. [3] Sardar, T. H., Khatun, A., Sengupta, S., Alam, Y., & Ara, T. (2024). Machine learning in the healthcare sector and the biomedical big data: Techniques, applications, and challenges. Big data computing, 336-352.
4. [4] Wani, N. A., Kumar, R., Bedi, J., & Rida, I. (2024). Explainable AI-driven IoMT fusion: Unravelling techniques, opportunities, and challenges with Explainable AI in healthcare. Information Fusion, 110, 102472.
5. [5] Hosain, M. T., Jim, J. R., Mridha, M. F., & Kabir, M. M. (2024). Explainable AI approaches in deep learning: Advancements, applications and challenges. Computers and electrical engineering, 117, 109246.
6. [6] Makumbura, R. K., Mampitiya, L., Rathnayake, N., Meddage, D. P. P., Henna, S., Dang, T. L., ... & Rathnayake, U. (2024). Advancing water quality assessment and prediction using machine learning models, coupled with explainable artificial intelligence (XAI) techniques like shapley additive explanations (SHAP) for interpreting the black-box nature. Results in Engineering, 23, 102831.
7. [7] Mignon, V. (2024). The multiple regression model. In Principles of Econometrics: Theory and Applications (pp. 105-170). Cham: Springer Nature Switzerland.
8. [8] Ramteke, N., & Maidamwar, P. (2023, July). Cardiac patient data classification using ensemble machine learning technique. In 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT) (pp. 1-6). IEEE.

9. [9] Ranjani, T., & Selvi, S. A. E. (2024, November). Comparative Analysis of ANN, XGBoost, and GridSearchCV-Tuned FNN for Diabetes Prediction. In 2024 2nd International Conference on Computing and Data Analytics (ICCDA) (pp. 1-5). IEEE.
10. [10] Ponce‐Bobadilla, A. V., Schmitt, V., Maier, C. S., Mensing, S., & Stodtmann, S. (2024). Practical guide to SHAP analysis: Explaining supervised machine learning model predictions in drug development. Clinical and translational science, 17(11), e70056.
11. [11] Lv, J., Chen, C. M., Kumari, S., & Li, K. (2025). Resource allocation for AI-native healthcare systems in 6G dense networks using deep reinforcement learning. Digital Communications and Networks.
12. [12] Wang, C., Divakarachari, P. B., & Jiang, H. (2025). 6G-Enabled Intelligent Healthcare Transport Systems: Framework and Resource Allocation Strategy. IEEE Transactions on Intelligent Transportation Systems.
13. [13] Alhussien, N., & Gulliver, T. A. (2024). Toward AI-enabled green 6G networks: A resource management perspective. IEEe Access, 12, 132972-132995.
14. [14] Saeed, M. M., Saeed, R. A., Abdelhaq, M., Alsaqour, R., Hasan, M. K., & Mokhtar, R. A. (2023). Anomaly detection in 6G networks using machine learning methods. Electronics, 12(15), 3300.
15. [15] Vincent, A. C. S. R., & Sengan, S. (2024). Edge computing-based ensemble learning model for health care decision systems. Scientific Reports, 14(1), 26997.
16. [16] Kumar, R., Madan, P., Shrivastava, A., Kumar, C. P., Rao, A. L. N., & Sankhyan, A. (2023, December). Ensemble-Based Big Data Analytics for Disease Prediction in Iot. In 2023 International Conference on Artificial Intelligence for Innovations in Healthcare Industries (ICAIIHI) (Vol. 1, pp. 1-6). IEEE.
17. [17] Gebreyesus, Y., Dalton, D., Nixon, S., De Chiara, D., & Chinnici, M. (2023). Machine learning for data center optimizations: feature selection using Shapley additive exPlanation (SHAP). Future Internet, 15(3), 88.
18. [18] Nohara, Y., Matsumoto, K., Soejima, H., & Nakashima, N. (2022). Explanation of machine learning models using shapley additive explanation and application for real data in hospital. Computer Methods and Programs in Biomedicine, 214, 106584.
19. [19] Greco, L., Luta, G., & Wilcox, R. (2024). On testing the equality between interquartile ranges. Computational statistics, 39(5), 2873-2898.
20. [20] Yağcıoğlu, M. (2025). A Comparative Study of Machine Learning Regression Models with and Without Dimensionality Reduction for Predicting Throughput in 5G Networks. Wireless Personal Communications, 143(1), 129-155.
21. [21] Yu, L., Zhou, R., Chen, R., & Lai, K. K. (2022). Missing data preprocessing in credit classification: One-hot encoding or imputation?. Emerging Markets Finance and Trade, 58(2), 472-482.
22. [22] Lin, G., Lin, A., & Gu, D. (2022). Using support vector regression and K-nearest neighbors for short-term traffic flow prediction based on maximal information coefficient. Information Sciences, 608, 517-531.
23. [23] Lu, H. W., & Lee, C. Y. (2021). Kernel-based dynamic ensemble technique for remaining useful life prediction. IEEE Robotics and Automation Letters, 7(2), 1142-1149.
24. [24] Acito, F. (2023). Predictive analytics with KNIME. Analytics for citizen data scientists. Switzerland: Springer.
25. [25] Rodriguez-Galiano, V., Sanchez-Castillo, M., Chica-Olmo, M., & Chica-Rivas, M. J. O. G. R. (2015). Machine learning predictive models for mineral prospectivity: An evaluation of neural networks, random forest, regression trees and support vector machines. Ore geology reviews, 71, 804-818.
26. [26] Bentéjac, C., Csörgő, A., & Martínez-Muñoz, G. (2021). A comparative analysis of gradient boosting algorithms. Artificial Intelligence Review, 54(3), 1937-1967.
27. [27] Kavzoglu, T., & Teke, A. (2022). Advanced hyperparameter optimization for improved spatial prediction of shallow landslides using extreme gradient boosting (XGBoost). Bulletin of Engineering Geology and the Environment, 81(5), 201.
28. [28] Meghanadha Reddy, A., Narendra Kumar, B., & Chatterjee, S. (2025). A novel kernel-based machine learning approach for phase analysis in modified sustainable concrete: Comparative insights from SVR and GPR on XRD data. Asian Journal of Civil Engineering, 26(12), 5317-5334.
29. [29] Camelia, T. S., Fahim, F. R., & Anwar, M. M. (2025, February). Optimizing 5G Quality of Service Using Machine Learning Models: A Comparative Analysis of MLR, SVR, and KNN Regression. In 2025 International Conference on Electrical, Computer and Communication Engineering (ECCE) (pp. 1-6). IEEE.
30. [30] Ahmad, M. N., Shao, Z., Xiao, X., Fu, P., Javed, A., & Ara, I. (2024). A novel ensemble learning approach to extract urban impervious surface based on machine learning algorithms using SAR and optical data. International Journal of Applied Earth Observation and Geoinformation, 132, 104013.
31. [31] Iakovlev, A. U., & Utochkin, I. S. (2023). Ensemble averaging: What can we learn from skewed feature distributions?. Journal of Vision, 23(1), 5-5.
32. [32] Dey, R., & Mathur, R. (2023, May). Ensemble learning method using stacking with base learner, a comparison. In International conference on data analytics and insights (pp. 159-169). Singapore: Springer Nature Singapore.
33. [33] Hasan, M., Abedin, M. Z., Hajek, P., Coussement, K., Sultan, M. N., & Lucey, B. (2025). A blending ensemble learning model for crude oil price forecasting. Annals of Operations Research, 353(2), 485-515.
34. [34] Yağcıoğlu, M. (2025). Machine learning based dynamic resource sharing and frequency reuse in 5G hetnets with dronecells. Computer Networks, 258, 111046.