Journals / Sinop Üniversitesi Fen Bilimleri Dergisi / 2026 / Cilt: 11 Sayı: 1

A Novel Triage Management for Digital Twin controlled Healthcare System

Dijital İkiz Kontrollü Sağlık Sistemi için Yeni Triyaj Yönetimi

Muhammed Gider(MANİSA CELÂL BAYAR ÜNİVERSİTESİ)0000-0002-0854-7479
Muge Erel-ozcevik(MANİSA CELÂL BAYAR ÜNİVERSİTESİ)0000-0003-3077-160X
Pages
296–320
Publication Date
2026-06-26
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The obtained results indicate that the proposed system provides an effective framework in terms of both data-driven classification and operational efficiency.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:The healthcare sector is undergoing a rapid transformation with technological advancements, and this transformation increases patients’ demand for faster, more effective, and more personalized healthcare services.
📊
Key Findings & Data:However, in emergency departments, problems such as insufficient resource management, inaccurate triage decisions, and consequently long waiting times reduce the effectiveness of healthcare services.
💡
Conclusion & Impact:The obtained results indicate that the proposed system provides an effective framework in terms of both data-driven classification and operational efficiency.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Triaj Yönetimi · Kuyruk Teorisi · Dijital İkiz

Abstract

The healthcare sector is undergoing a rapid transformation with technological advancements, and this transformation increases patients’ demand for faster, more effective, and more personalized healthcare services. However, in emergency departments, problems such as insufficient resource management, inaccurate triage decisions, and consequently long waiting times reduce the effectiveness of healthcare services. Therefore, there is a need for new triage approaches that can reduce patient congestion and waiting times in emergency departments and improve resource utilization. In the literature, there are many different studies using artificial intelligence or queueing theory, and according to these studies, such next-generation approaches provide certain gains in improving patient waiting times. In this study, unlike the existing literature, a next-generation healthcare system is proposed with the aim of reducing congestion in emergency departments, minimizing waiting times, and monitoring patients’ health status in real time through an artificial intelligence-based model, the analytical capability of queueing theory, and efficient resource management. The integration of this new structure is carried out under the control of a digital twin. In this context, artificial intelligence is used as a decision-support mechanism in determining triage priority by jointly evaluating patients’ real-time and historical health data; queueing theory, on the other hand, provides analytical evaluations regarding patient flow, waiting time, service capacity, and resource utilization, thereby improving the operational efficiency of the system. A next-generation triage management system is proposed, starting with patient identification and the measurement of real-time health parameters. By using instantaneously collected and historical health data, the future condition of the patient is predicted with the support of the Digital Twin. Here, the proposed new triage management algorithm performs green, yellow, or red classification. Iterative parameter optimization was applied to the algorithm, and waiting-time performance was evaluated under different scenarios. In this framework, hospital capacity is appropriately assessed through theoretical analyses based on queueing theory, and personnel and resource management are optimized. As a result of the simulation tests, it was observed that the proposed system reduced patient waiting times and kept them below 30 minutes. AutoGluon-based classification experiments demonstrated that data balancing strategies were decisive in triage classification performance. In particular, random oversampling and ADASYN + Tomek Links approaches contributed to a more consistent separation of red, yellow, and green patient groups by reducing class imbalance. In addition, MATLAB/Simulink-based queueing theory analyses showed that the average waiting times were approximately 29, 14, and 7 minutes for the green, yellow, and red categories, respectively. The obtained results indicate that the proposed system provides an effective framework in terms of both data-driven classification and operational efficiency.

Özet

Sağlık sektörü, teknolojik gelişmelerle birlikte hızlı bir dönüşüm geçirmekte ve bu dönüşüm, hastaların daha hızlı, etkili ve kişiselleştirilmiş sağlık hizmetlerine erişim talebini artırmaktadır. Ancak acil servislerde; yetersiz kaynak yönetimi, yanlış triyaj kararları ve bu nedenle uzun bekleme süreleri gibi sorunlar sağlık hizmetlerinin etkinliğini azaltmaktadır. Bu nedenle, acil servislerdeki bekleyen hasta yoğunluğunu ve bekleme sürelerini azaltabilecek ve kaynak kullanımını iyileştirebilecek yeni triyaj yaklaşımlarına ihtiyaç bulunmaktadır. Literatürde, yapay zeka veya kuyruk teorisi kullanan birçok farklı çalışma bulunmaktadır ve bu çalışmalara göre bu yeni nesil yaklaşımların hasta bekleme sürelerini iyileştirmede belirli kazanımlar sağladığı belirtilmektedir. Bu çalışmada ise literatürden farklı olarak, yapay zeka temelli model ve kuyruk teorisi tabanlı analitik analiz yeteneği ile verimli kaynak yönetimi sayesinde acil servislerdeki yoğunluğun azaltılması, bekleme sürelerinin minimize edilmesi ve hastaların sağlık durumlarının anlık olarak takibi amaçlanarak yeni nesil bir sağlık sistemi sunulmaktadır. Bu yeni yapının entegrasyonu ise dijital ikiz kontrolünde gerçekleştirilmektedir. Bu kapsamda yapay zekâ, hastaların anlık ve geçmiş sağlık verilerini birlikte değerlendirerek triyaj önceliğinin belirlenmesinde karar destek mekanizması olarak kullanılmakta kuyruk teorisi ise hasta akışı, bekleme süresi, hizmet kapasitesi ve kaynak kullanımına ilişkin analitik değerlendirmeler sağlayarak sistemin operasyonel verimliliğini artırmaktadır. Hastaların kimlik tespiti ve anlık sağlık parametrelerinin ölçülmesiyle başlayan yeni nesil bir triyaj yönetim sistemi önerilmektedir. Anlık alınan ve geçmiş sağlık verileri kullanılarak Dijital İkiz desteği ile hastanın gelecekteki durumu tahminlenir. Burada önerilen yeni triyaj yönetimi algoritması ile yeşil, sarı veya kırmızı sınıflandırması gerçekleştirilir. Algoritma üzerinde iteratif parametre optimizasyonu uygulanmış ve farklı senaryolarda bekleme süresi performansı değerlendirilmiştir. Burada, kuyruk teorisinden faydalanılarak teori kanalizler yardımıyla hastane kapasitesi uygun bir şekilde değerlendirilmekte ve personel ile kaynak yönetimi optimize edilmektedir. modelin simülasyon testleri sonucunda, önerilen sistemin hasta bekleme sürelerini azaltarak süreleri 30 dakikanın altında tuttuğu görülmüştür. AutoGluon tabanlı sınıflandırma deneyleri ,veri dengeleme stratejilerinin triyaj sınıflandırma başarımı üzerinde belirleyici olduğunu göstermiştir. Özellikle random over sampling ve ADASYN+Tomek Links yaklaşımları, sınıf dengesizliğini azaltarak kırmızı, sarı ve yeşil hasta gruplarının daha tutarlı biçimde ayrıştırılmasına katkı sağlamıştır. Buna ek olarak, MATLAB/Simulink tabanlı kuyruk teorisi analizlerinde ortalama bekleme süreleri yeşil, sarı ve kırmızı kategoriler için sırasıyla yaklaşık 29, 14 ve 7dakika olarak hesaplanmıştır.

🔖

Cite This Article

Gider, M., Erel-ozcevik, M. (2026). A Novel Triage Management for Digital Twin controlled Healthcare System. Sinop Üniversitesi Fen Bilimleri Dergisi, Cilt: 11 Sayı: 1, 296–320. https://doi.org/10.33484/sinopfbd.1764220

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (21)

  1. Sağlık Bakanlığı, T. C. (2023). Covid-19 pandemi sürecine sağlık sisteminin uyumu: Türkiye'deki yoğun bakımlar (tedavi, süreçler, altyapı ve insan gücü), teknik rapor. https://dosyamerkez.saglik.gov.tr/Eklenti/47833/0/tryogunbakimstdr05032024pdf.pdf
  2. Penverne, Y., Martinez, C., Cellier, N., Pehlivan, C., Jenvrin, J., Savary, D., Debierre, V., Deciron, F., Bichri, A., Lebastard, Q., Montassier, E., Leclere, B., & Fontanili, F. (2024). A simulation based digital twin approach to assessing the organization of response to emergency calls. npj Digital Medicine, 7, 385. https://doi.org/10.1038/s41746-024-01392-2
  3. Schiza, E. C., Kyprianou, T. C., Petkov, N., & Schizas, C. N. (2019). Proposal for an ehealth based ecosystem serving national healthcare. IEEE Journal of Biomedical and Health Informatics, 23(3), 1346-1357. https://doi.org/10.1109/JBHI.2018.2834230
  4. Bahiddin, N., Othman, Z. A., Bakar, A. A., & Rahim, N. A. A. (2022). An interrelated decision-making model for an intelligent decision support system in healthcare. IEEE Access, 10, 31660-31676. https://doi.org/10.1109/ACCESS.2022.3160725
  5. Chengyu, Y. (2024). U-sort: Utah support for emergency operating room triage. Technical Report UUCS-24-010, School of Computing, University of Utah. https://www-old.cs.utah.edu/docs/techreports/2024/UUCS-24-010.pdf
  6. Chughtai, S. (2024). Ai-empowered emergency department cutting edge digital medicine. Journal of Pakistan Society of Internal Medicine, 5(4), 758-761. https://doi.org/10.70302/jpsim.v5i4.2481
  7. Al-Hawawreh, M., & Hossain, M. S. (2024). Digital twin-driven secured edge-private cloud industrial internet of things (iiot) framework. Journal of Network and Computer Applications, 226, 103888. https://doi.org/10.1016/j.jnca.2024.103888
  8. Li, Z., Xue, T., Yu, L., & Yang, B. (2023). Performance analysis of virtual-real synchronization in digital twin network: Trade-offs between timeliness, distortion, and sustainability. In 2023 3rd International Conference on Intelligent Communications and Computing (ICC), 289-295. https://doi.org/10.1109/ICC59986.2023.10420950
  9. Al Sawidan, S. A. M., Alsalah, A. J. M., Alsalah, B. J. M., Abosaaq, A. J., Alalhareth, N. D., Mohammed Al Swidan, A. M. M., Alyami, H. A. H., Al Mansour, A. S. M., Aldosari, A. A. S., & Almas, Y. H. S. (2024). Optimizing prehospital stroke care: A comprehensive literature review. Journal of Ecohumanism, 3(8), 1769-1775. https://doi.org/10.62754/joe.v3i8.4866
  10. Tahernejad, A., Sahebi, A., Salehi Sahl Abadi, A., & Safari, M. (2024). Application of artificial intelligence in triage in emergencies and disasters: a systematic review. BMC Public Health, 24(1), 3203. https://doi.org/10.1186/s12889-024-20447-3
  11. Magalhães, M. E. A., da Silva, C. V. L., de Oliveira, H. M., do Nascimento, G. L. S., de Lima, A. B. R., Flores, M. T. S., Leite, I. F., da Silva, G. A., de Faria, I. A. F. K., da Cruz, A. N., Costa, J. H. d. C., & Zanoni, R. D. (2024). The use of artificial intelligence in patient triage in emergency departments: an integrative review. Revista de Gestão Social e Ambiental, 18(12), 1-12. https://doi.org/10.24857/rgsa.v18n12-052
  12. Sulaiman, W. A., Stylianides, C., Nikolaou, A., Antoniou, Z., Constantinou, I., Palazis, L., Vavlitou, A., Kyprianou, T., Kyriacou, E., Kakas, A., Pattichis, M. S., Panayides, A. S., & Pattichis, C. S. (2025). Leveraging machine learning and rule extraction for enhanced transparency in emergency department length of stay prediction. Front Digit Health, 6. https://doi.org/10.3389/fdght.2024.1498939
  13. Olivia, D., Attigeri, G., & Saxena, A. (2024). Optimization model for mass casualty management system using qos-aware routing protocol and casualty triage prediction. International Journal of Information Technology. https://doi.org/10.1007/s41870-024-02052-0
  14. Ammari, S., Camez, A. O., Ayobi, A., Quenet, S., Zemmour, A., Mniai, E. M., Chaibi, Y., Franciosini, A., Clavel, L., Bidault, F., Muller, S., Lassau, N., Balleyguiert, C., & Assi, T. (2024). Contribution of an artificial intelligence tool in the detection of incidental pulmonary embolism on oncology assessment scans. Life, 14(11). https://doi.org/10.3390/life14111347
  15. Pradhan, B., Das, S., Roy, D. S., Routray, S., Benedetto, F., & Jhaveri, R. H. (2023). An ai-assisted smart healthcare system using 5g communication. IEEE Access. https://doi.org/10.1109/ACCESS.2023.3317174
  16. Lee, H. K., Musa, A. J., Bain, P. A., Nelson, K., Baker, C., & Li, J. (2019). A queueing network model for analysis of patient transitions within hospitals. IEEE Transactions on Automation Science and Engineering, 16(1), 6-20. https://doi.org/10.1109/TASE.2018.2793251
  17. Ferreira, L., Pilastri, A., Martins, C. M., Pires, P. M., & Cortez, P. (2021). A comparison of automl tools for machine learning, deep learning and xgboost. 2021 International Joint Conference on Neural Networks (IJCNN), 1-8. https://doi.org/10.1109/IJCNN52387.2021.9534091
  18. Nhita, F., Adiwijaya, & Kurniawan, I. (2023). Improvement of imbalanced data handling: A hybrid sampling approach by using adaptive synthetic sampling and tomek links. 2023 Eighth International Conference on Informatics and Computing (ICIC), 1-5. https://doi.org/10.1109/ICIC60109.2023.10381929
  19. Sun, Z., Zhang, Z., Chen, Y., Liu, S., & Song, Y. (2020). Frost filtering algorithm of sar images with adaptive windowing and adaptive tuning factor. IEEE Geoscience and Remote Sensing Letters, 17(6), 1097-1101. https://doi.org/10.1109/LGRS.2019.2939208
  20. Yadav, S., Maravi, Y. P. S., Agrawal, J. w., & Mishra, N. (2021). A neural network based diabetes prediction on imbalanced data. 2021 10th IEEE International Conference on Communication Systems and Network Technologies (CSNT). https://doi.org/10.1109/CSNT51715.2021.9509732
  21. Tasin, I., Nabil, T. U., Islam, S., & Khan, R. (2023). Diabetes prediction using machine learning and explainable ai techniques. Healthcare Technology Letters, 10, 1-10. https://doi.org/10.1049/htl2.12039