Journals / Kocatepe Tıp Dergisi / 2026 / Cilt: 27 Sayı: 1

COMPARATIVE ANALYSIS OF MACHINE LEARNING VERSUS CONVENTIONAL MODELS FOR PREDICTING ERYTHROCYTE NEED IN CARDIOVASCULAR SURGERY

KARDİYOVASKÜLER CERRAHİDE ERITROSİT İHTİYACINI TAHMİN ETMEK İÇİN MAKİNE ÖĞRENMESİ İLE GELENEKSEL MODELLERİN KARŞILAŞTIRMALI ANALİZİ

Ahmet Yuksek(Kocaeli Şehir Hastanesi)0000-0002-7529-2971
İpek Yakın Düzyol(Kocaeli Şehir Hastanesi)0000-0003-0374-8121
Gamze Talih(ERCİYES ÜNİVERSİTESİ, TIP FAKÜLTESİ)0000-0003-4743-9734
Pages
42–48
Publication Date
2026-01-12
🏛️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 ability of ML algorithms to reinforce themselves with local data and adapt to changing clinical contexts highlights their potential for enhancing predictive accuracy over time.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:OBJECTIVE: Excessive bleeding during cardiovascular surgery poses significant risks to patients, necessitating accurate prediction of erythrocyte suspension (ES) requirements for optimal perioperative management.
📊
Key Findings & Data:RESULTS: A total of 430 patients undergoing isolated CABG surgeries were included in the analysis.
💡
Conclusion & Impact:The ability of ML algorithms to reinforce themselves with local data and adapt to changing clinical contexts highlights their potential for enhancing predictive accuracy over time.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Makine Öğrenmesi · Kardiyovasküler Cerrahi · Perioperatif Kanama

Abstract

OBJECTIVE: Excessive bleeding during cardiovascular surgery poses significant risks to patients, necessitating accurate prediction of erythrocyte suspension (ES) requirements for optimal perioperative management. This study aims to compare the efficacy of machine learning (ML) algorithms against conventional scorings in predicting ES needs during coronary artery bypass grafting (CABG) surgeries. MATERIAL AND METHODS: The ML algorithm, developed using patient demographic, preoperative, and surgical data, was compared with six established bleeding prediction scores. ES need predictions were analyzed using logistic regression-based neural networks. RESULTS: A total of 430 patients undergoing isolated CABG surgeries were included in the analysis. ML algorithms demonstrated comparable predictive power to conventional scoring systems, with an accuracy of 75.5% for predicting any ES need and 93.8% for predicting the need for >2 units of ES. Among traditional scoring systems, the TRUST score exhibited the highest predictive ability, followed by the TRACK and WILL BLEED scores. CONCLUSIONS: The ML algorithm showed potential for local adaptation and self-improvement, enhancing its utility in perioperative planning. Our findings underscore the feasibility of ML algorithms in optimizing blood product utilization in cardiovascular surgery. Further research is warranted to explore the scalability and generalizability of these models across diverse clinical settings. ML predictions for ES transfusion readiness in patients prior to CABG surgery demonstrate a high degree of accuracy and comparable performance to traditional scoring systems. The ability of ML algorithms to reinforce themselves with local data and adapt to changing clinical contexts highlights their potential for enhancing predictive accuracy over time.

Özet

AMAÇ: Kardiyovasküler cerrahi sırasında aşırı kanama, hastalar için önemli riskler oluşturur ve optimal perioperatif yönetim için eritrosit süspansiyonu (ES) gereksinimlerinin doğru bir şekilde tahmin edilmesini gerektirir. Bu çalışma, koroner arter baypas greftleme (CABG) ameliyatları sırasında ES ihtiyaçlarını tahmin etmede makine öğrenimi (ML) algoritmalarının geleneksel puanlamalara karşı etkinliğini karşılaştırmayı amaçlamaktadır. GEREÇ VE YÖNTEM: Hastaların demografik, ameliyat öncesi ve cerrahi verileri kullanılarak geliştirilen ML algoritması, altı bilinen kanama tahmin skoruyla karşılaştırıldı. ES ihtiyacı tahminleri lojistik regresyon tabanlı sinir ağları kullanılarak analiz edildi. BULGULAR: İzole CABG ameliyatları geçiren toplam 430 hasta analize dahil edildi. ML algoritmaları, herhangi bir ES ihtiyacını tahmin etmede %75,5 ve >2 ünite ES ihtiyacını tahmin etmede %93,8 doğrulukla geleneksel puanlama sistemlerine benzer tahmin gücü gösterdi. Geleneksel puanlama sistemleri arasında, TRUST puanı en yüksek öngörü yeteneğini sergiledi, ardından TRACK ve WILL BLEED puanları geldi. SONUÇ: ML algoritması, perioperatif planlamada kullanımını artırarak, yerel adaptasyon ve kendini iyileştirme potansiyeli gösterdi. Bulgularımız, kardiyovasküler cerrahide kan ürünü kullanımını optimize etmede ML algoritmalarının uygulanabilirliğini vurgulamaktadır. Bu modellerin çeşitli klinik ortamlarda ölçeklenebilirliğini ve genelleştirilebilirliğini keşfetmek için daha fazla araştırma yapılması gerekmektedir. CABG ameliyatından önce hastalarda ES transfüzyon hazırlığı için ML tahminleri, geleneksel puanlama sistemlerine kıyasla yüksek derecede doğruluk ve karşılaştırılabilir performans göstermektedir. ML algoritmalarının kendilerini yerel verilerle güçlendirme ve değişen klinik bağlamlara uyum sağlama yeteneği, zaman içinde öngörü doğruluğunu artırma potansiyellerini vurgulamaktadır.

🔖

Cite This Article

Yuksek, A., Düzyol, İ. Y., Talih, G. (2026). COMPARATIVE ANALYSIS OF MACHINE LEARNING VERSUS CONVENTIONAL MODELS FOR PREDICTING ERYTHROCYTE NEED IN CARDIOVASCULAR SURGERY. Kocatepe Tıp Dergisi, Cilt: 27 Sayı: 1, 42–48. https://doi.org/10.18229/kocatepetip.1626376

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (24)

  1. Al-Shammari F, Al-Duaij A, Al-Fadhl Ji, et al. Blood Component Transfusion in Primary Coronary Artery Bypass Surgery in Kuwait. Med Princ Pract. 2005;14(2):83–6.
  2. Erdoes G, Faraoni D, Koster A, et al. Perioperative Considerations in Management of the Severely Bleeding Coagulopathic Patient. Anesthesiology. 2023;138(5):535-60.
  3. Liu Y, Wang X, Chen ZY, Zhang WL, et al. Severe bleeding following off-pump coronary artery bypass grafting: predictive factors and risk model. J Geriatr Cardiol. 2021;18(6):449-61.
  4. Bartoszko J, Wijeysundera DN, Karkouti K, et al. Transfusion Avoidance in Cardiac Surgery Study Investigators. Comparison of Two Major Perioperative Bleeding Scores for Cardiac Surgery Trials: Universal Definition of Perioperative Bleeding in Cardiac Surgery and European Coronary Artery Bypass Grafting Bleeding Severity Grade. Anesthesiology. 2018;129(6):1092-100.
  5. Char DS, Burgart A. Machine-Learning Implementation in Clinical Anesthesia: Opportunities and Challenges. Anesth Analg. 2020;130(6):1709-12.
  6. Demirel E, Özer Gökaslan Ç. Radiomics Feature-Based Machine Learning For Preoperative Grading of Meningiomas: A Study Using Automl. Kocatepe Medical Journal. 2025;26(2):168-73.
  7. Kewal K. Jain. Personalized Management of Cardiovascular Disorders. Med Princ Pract. 2017;26 (5):399–414.
  8. Hever G, Cohen L, O'Connor MF, et al. Machine learning applied to multi-sensor information to reduce false alarm rate in the ICU. J Clin Monit Comput. 2020;34(2):339-52.
  9. Gao Y, Liu X, Wang L, et al. Machine learning algorithms to predict major bleeding after isolated coronary artery bypass grafting. Front Cardiovasc Med. 2022(28);9:881881.
  10. Salsano A, Dominici C, Nenna A, et al. Predictive scores for major bleeding after coronary artery bypass surgery in low operative risk patients. J Cardiovasc Surg (Torino). 2020;61(2):234-42.
  11. Vlot EA, Vernooij LM, Loer SA, et al. External Validation of the ACTA-PORT Transfusion Risk Score in Older Cardiac Surgery Patients at Risk of Frailty. J Cardiothorac Vasc Anesth. 2022;36(8):2983-90.
  12. Biancari F, Brascia D, Onorati F, et al. Prediction of severe bleeding after coronary surgery: the WILL-BLEED Risk Score. Thrombosis and Haemostasis. 2017;117(3):445-56.
  13. Günertem E, Urcun S, Pala AA, et al. Predictiveness of different preoperative risk assessments for postoperative bleeding after coronary artery bypass grafting surgery. Perfusion. 2021:36(3):277-84.
  14. Kim TS, Lee JH, An H, Na C-Y. Transfusion risk and clinical knowledge (TRACK) score and cardiac surgery in patients refusing transfusion. Journal of Cardiothoracic and Vascular Anesthesia. 2016;30(2):373-8.
  15. Alghamdi AA, Davis A, Brister S, et al. Development and validation of Transfusion Risk Understanding Scoring Tool (TRUST) to stratify cardiac surgery patients according to their blood transfusion needs. Transfusion. 2006;46(7):1120-9.
  16. Bento D, Marques N, Azevedo P, et al. CRUSADE: Is it still a good score to predict bleeding in acute coronary syndrome? Revista Portuguesa de Cardiologia (English Edition). 2018;37(11):889-97.
  17. Yavuz T, Büyükerkmen E, Doğan Bakı et al. Is Entry Hemoglobin Value An Intensive Care and Mortality The Determinant in Trauma Patients With Extremity and Pelvis Fracture?. Kocatepe Medical Journal. 2021;22(5):355- 9.
  18. Chegini A, Alaei M, Jahangiryan A, et al. Blood Products Utilization Status in Off-Pump Cardiac Surgery Patients. Int J Hematol Oncol Stem Cell Res. 2023(1);17(3):200-9.
  19. Whitney GM, Woods MC, France DJ, et al. Reducing intraoperative red blood cell unit wastage in a large Academic Medical Center. Transfusion. 2015;55(11):2752-8.
  20. Hassan AM, Biaggi AP, Asaad M, et al. Development and Assessment of Machine Learning Models for Individualized Risk Assessment of Mastectomy Skin Flap Necrosis. Ann Surg. 2023;278(1):123-30.
  21. Maciąg TT, van Amsterdam K, Ballast A, et al. Machine learning in anesthesiology: Detecting adverse events in clinical practice. Health Informatics Journal. 2022;28(3):14604582221112855.
  22. Silva GFS, Fagundes TP, Teixeira BC, et al. Machine Learning for Hypertension Prediction: a Systematic Review. Curr Hypertens Rep. 2022;24(11):523-33.
  23. Deo RC. Machine Learning in Medicine. Circulation. 2015;132(20):1920-30.
  24. Angelo D'Alessandro, Peter Bugert; Omics and Machine Learning in Transfusion Medicine. Transfus Med Hemother. 2023;50(3):161-2.