Yapay zeka algoritmalarının epitelyal over kanseri hastalarında progresyonsuz sağkalımı tahmin etmedeki etkinliği

Abstract

Amaç: Bu çalışmanın amacı, epitelyal over kanseri tanısı alan hastalarda progresyonsuz sağkalımı (PFS) tahmin etmede yapay zeka tabanlı modellerin öngörü performansını değerlendirmek ve yorumlanabilir makine öğrenimi yaklaşımlarının karşılaştırmalı analizini yapmaktır. Gereç ve Yöntem: Ocak 2015 ile Aralık 2020 tarihleri arasında Antalya Eğitim ve Araştırma Hastanesi Jinekolojik Onkoloji Anabilim Dalı'nda cerrahi müdahaleye alınan toplam 159 hasta retrospektif olarak çalışmaya dahil edildi. Sonuç analizi için klinik veriler kullanılmış ve kohort rastgele bir eğitim grubu (n = 117; %75) ve bir doğrulama grubu (n = 42; %25) olarak sınıflandırılmıştır. Çalışmamız için en popüler ve uygun olan sekiz algoritma modeli kullanılarak bir makine öğrenimi (ML) çalışması yapılmıştır. Modelin oluşturulması için gereken süre, ortalama mutlak hata, kök ortalama kare hata ve korelasyon katsayısı belirlendi. Sonuçlar: Rastgele Orman algoritması en başarılı algoritma olarak ortaya çıktı. Bu algoritma, sonraki araştırmaların odak noktası oldu. Rastgele Orman algoritmasının korelasyon katsayısı 0,5731, ortalama mutlak hatası 16,45 ve kök ortalama kare hatası 20,98 idi. Modeli oluşturmak için gereken süre 0,03 saniyeydi. Kalan algoritmalar Doğrusal Regresyon, Bootstrap Aggregating (bagging), Additive Regression, Random Committee ve Regression by Discretization (Korelasyon Katsayıları: 0,5326, 0,4915, 0,4491, 0,4077, 0,3817) idi. En iyi performans gösteren Rastgele Orman algoritması için gerçek ve tahmin edilen PFS değerleri arasında orta düzeyde bir korelasyon gözlemlenmiştir (Korelasyon Katsayısı: 0,4-0,6), bu da tahminlerde orta düzeyde bir başarı oranına işaret etmektedir.

Keywords

• Yapay zeka
• Jinekolojik onkolojide derin öğrenme
• Epiteliyal over neoplazmları

Effectiveness of artificial intelligence algorithms in predicting progression-free survival in epithelial ovarian cancer patients

Abstract

Aims: This study aimed to assess the predictive performance of artificial intelligence–based models in estimating progressionfree survival (PFS) in patients with epithelial ovarian cancer and to compare various interpretable machine learning approaches. Methods: Between January 2015 and December 2020, a total of 167 patients who underwent surgical intervention at the Gynaecological Oncology Department of Antalya Training and Research Hospital were retrospectively included in the study if their data were complete. Clinical data were analysed, and the dataset was randomly divided into a training group (n=117; 75%) and a validation group (n=42; 25%). A machine learning (ML) analysis was conducted using the eight most relevant and widely applied algorithmic models for this study design. Model development time, mean absolute error (MAE), root mean square error (RMSE), and correlation coefficient (CC) were evaluated. Results: Random Forest demonstrated the highest accuracy (MAE=16.45, CC=0.571, RMSE=20.98, time=0.03) and thus became the focus of subsequent analyses. Other algorithms included Linear Regression, Bootstrap Aggregating, Additive Regression, Random Committee, and Regression by Discretization (CC=0.533, 0.492, 0.449, 0.408, and 0.382, respectively). For Random Forest, a moderate correlation was observed between actual and predicted PFS values (CC=0.4–0.6), indicating moderate predictive performance. Conclusion: The findings of this study demonstrate that machine learning models, particularly Random Forest, can achieve moderate yet clinically relevant prognostic performance based on routinely collected clinical data. In particular, Random Forest demonstrates potential clinical value in guiding patient follow-up strategies and supporting individualized management in ovarian cancer, although further research is required to enhance its clinical validity and applicability.

Keywords

• Artificial intelligence
• deep learning in gynecologic oncology
• epithelial ovarian neoplasms

References

1. Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209-249. doi:10. 3322/caac.21660
2. Shachar E, Raz Y, Rotkop G, et al. Cytoreductive surgery in advanced epithelial ovarian cancer: a real-world analysis guided by clinical variables, homologous recombination, and BRCA status. Int J Gynecol Cancer. 2025;35(6):101809. doi:10.1016/j.ijgc.2025.101809
3. Caruso G, Tomao F, Parma G, et al. Poly (ADP-ribose) polymerase inhibitors (PARPi) in ovarian cancer: lessons learned and future directions. Int J Gynecol Cancer. 2023;33(4):431-443. doi:10.1136/ijgc-2022-004149
4. Atallah GA, Kampan NC, Chew KT, et al. Predicting prognosis and platinum resistance in ovarian cancer: role of immunohistochemistry biomarkers. Int J Mol Sci. 2023;24(3):1973. doi:10.3390/ijms24031973
5. Cruz JA, Wishart DS. Applications of machine learning in cancer prediction and prognosis. Cancer Inform. 2006;2:117693510600200030. doi:10.1177/117693510600200030
6. Fountzilas E, Pearce T, Baysal MA, Chakraborty A, Tsimberidou AM. Convergence of evolving artificial intelligence and machine learning techniques in precision oncology. NPJ Digit Med. 2025;8(1):75. doi:10. 1038/s41746-025-01471-y
7. Abbas S, Asif M, Rehman A, Alharbi M, Khan MA, Elmitwally N. Emerging research trends in artificial intelligence for cancer diagnostic systems: a comprehensive review. Heliyon. 2024;10(17):e36743. doi:10. 1016/j.heliyon.2024.e36743
8. Enshaei A, Robson C, Edmondson R. Artificial intelligence systems as prognostic and predictive tools in ovarian cancer. Ann Surg Oncol. 2015; 22:3970-3975. doi:10.1245/s10434-015-4475-6

9. Krzyszczyk P, Acevedo A, Davidoff EJ, et al. The growing role of precision and personalized medicine for cancer treatment. Technol Cancer Res Treat. 2018;6(03n04):79-100. doi:10.1142/S2339547818300020
10. Fagotti A, Ferrandina MG, Vizzielli G, et al. Randomized trial of primary debulking surgery versus neoadjuvant chemotherapy for advanced epithelial ovarian cancer (SCORPION-NCT01461850). Int J Gynecol Cancer. 2020;30(11):1657-1664. doi:10.1136/ijgc-2020-001640
11. Van Meurs HS, Tajik P, Hof MH, et al. Which patients benefit most from primary surgery or neoadjuvant chemotherapy in stage IIIC or IV ovarian cancer? An exploratory analysis of the European Organisation for Research and Treatment of Cancer 55971 randomised trial. Eur J Cancer. 2013;49(15):3191-3201. doi:10.1016/j.ejca.2013.06.013
12. Clavien PA, Barkun J, De Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg. 2009;250(2):187-196. doi:10.1097/SLA.0b013e3181b13ca2
13. Biau G, Scornet E. A random forest guided tour. Test. 2016;25(2):197-227.
14. Fernández-Delgado M, Cernadas E, Barro S, Amorim D. Do we need hundreds of classifiers to solve real world classification problems? J Machine Learning Res. 2014;15(1):3133-3181.
15. Jian L, Chen X, Hu P, et al. Predicting progression-free survival in patients with epithelial ovarian cancer using an interpretable random forest model. Heliyon. 2024;10(15):e35344. doi:10.1016/j.heliyon.2024.e 35344
16. Laios A, Katsenou A, Tan YS, et al. Feature selection is critical for 2-year prognosis in advanced stage high grade serous ovarian cancer by using machine learning. Cancer Control. 2021;28:10732748211044678. doi:10. 1177/10732748211044678
17. Piedimonte G, Fusco R, Rizzo A, et al. Machine learning models based on radiomics for the prediction of ovarian cancer outcomes: a systematic review. Cancers (Basel). 2025;17(3):336. doi:10.3390/cancers17030336
18. Maiorano BA, Rabaiotti E. Prognostic and predictive potential of radiogenomics in epithelial ovarian cancer: a systematic review and meta-analysis. Onco. 2025;6(4):84. doi:10.3390/onco6040084
19. Zeng S, Wang XL, Yang H. Radiomics and radiogenomics: extracting more information from medical images for the diagnosis and prognostic prediction of ovarian cancer. Mil Med Res. 2024;11(1):77. doi:10.1186/s40779-024-00580-1
20. Wu J, Zhao L, Li Y, et al. AIDPI: a transcriptomic and multi-omics prognostic model for ovarian cancer. Front Genet. 2025;16:12268085.
21. Chen X, Liu Y, Zhang H, et al. Construction and validation of a CSOARG gene signature for prognosis in ovarian cancer. Front Oncol. 2025;15:1592426. doi:10.3389/fonc.2025.1592426
22. Jiang Y, Yang C, Xu W, et al. AUTOSurv: deep learning framework integrating multi-omics and clinical data for ovarian cancer survival prediction. NPJ Precis Oncol. 2024;8:37. doi:10.1038/s41698-023-00494-6
23. Laios A, Kalampokis E, Johnson R, et al. Explainable artificial intelligence for prediction of complete surgical cytoreduction in advanced-stage epithelial ovarian cancer. J Pers Med. 2022;12(4):607. doi:10.3390/jpm12040607
24. Harrell J, Frank E, Harrell FE. Cox proportional hazards regression model. In: Regression modeling strategies: with applications to linear models, logistic and ordinal regression, and survival analysis. 2nd ed. Springer; 2015. doi:10.1007/978-3-319-19425-7
25. Bertolaccini L, Pardolesi A, Davoli F, Solli P. Nanos gigantium humeris insidentes: the awarded Cox proportional hazards model. J Thorac Dis. 2016;8(11):3464-3470. doi:10.21037/jtd.2016.11.84
26. Jennings B, Zhang T. A systematic review of multimodal AI models for ovarian cancer prognosis using whole slide imaging and omics data. arXiv preprint. 2025;arXiv:2507.16876.