Diferansiye tiroid kanseri nüksünün açıklanabilir yapay zeka modeli ile tahmin edilmesi

Year 2025, Volume: 6 Issue: 3, 280 - 287, 28.09.2025

Ahmet Cankat Öztürk , Erkan Akkur , Serkan Çizmecioğullari

https://doi.org/10.47482/acmr.1677545

Abstract

Amaç: Bu çalışmanın amacı, diferansiye tiroid kanserinin nüksünü tahmin etmek ve açıklanabilir yapay zeka modeli aracılığıyla nüksün en temsili özelliklerini belirlemektir.
Gereç ve yöntemler: Veri kümesi UCI Makine Öğrenimi Deposundan elde edilmiş ve “Diferansiye Tiroid Kanseri Nüksü” olarak adlandırılmıştır. Bu veri kümesi 383 hasta ve 17 özellik içermektedir. Diferansiye tiroid kanserinin nüksünü tahmin etmek için Random Forest, Gradient Boosting, AdaBoost, Support Vector Classifier ve Logistic Regression olmak üzere beş sınıflandırıcı kullanılmıştır. Tahmin sonucu üzerinde en fazla etkiye sahip özellikleri belirlemek için permütasyon özellik önemi ve Shapley açıklanabilir yapay zeka yöntemleri kullanılmıştır.
Sonuçlar: Random Forest algoritması sırasıyla %0,9739 ve 0,993 ile en yüksek doğruluk ve AUC-ROC skorunu elde etmiştir. En belirgin risk faktörleri tedavi yanıtı, ATA riski, tümör ve yaş olarak belirlenmiştir.
Sonuç: Önerilen açıklanabilir makine öğrenimi modeli, diferansiye tiroid kanserinin nüksünü tahmin etmek ve en belirgin özellikleri belirlemek için güvenilir bir model olarak kabul edilebilir. Bu model, sağlık profesyonellerine yardımcı olma potansiyeline sahip olabilir.

Keywords

Diferansiye tiroid kanseri , nüks , makine öğrenmesi , açıklanabilir yapay zeka , shapley , permütasyon özellik önemi

Predicting recurrence of differentiated thyroid cancer with an explainable artificial intelligence model

Abstract

Background: This study aimed to predict the recurrence of differentiated thyroid cancer and identify its most representative risk factors using an explainable artificial intelligence model.
Methods: The publicly available Differentiated Thyroid Cancer Recurrence dataset from the University of California Irvine Machine Learning Repository, comprising 383 patients and 17 features, was employed. Five classifiers, -Random Forest, Gradient Boosting, AdaBoost, Support Vector Classifier and Logistic Regression-, were employed to predict the recurrence. Permutation feature importance (PFI) and SHapley Additive exPlanations (SHAP) explainable artificial intelligence methods were used to determine the features that had the most impact on the prediction result.
Results: The Random Forest algorithm outperformed others, achieving an accuracy of 97.39% and an Area under the Curve of 0.993. Response to treatment, ATA risk stratification, tumor stage and patient age were determined as the factors with the highest contribution to the model prediction process through SHAP and permutation importance analyses, and this finding was consistent with the prognostic markers stated in the literature.
Conclusion: The proposed explainable machine learning framework has shown satisfactory results in predicting DTC recurrence while identifying clinically important features. This approach can offer valuable support to clinicians in early identification of high-risk patients and personalization of surveillance strategies.

Keywords

Differentiated thyroid cancer , recurrence , machine learning , explainable artificial intelligence , shapley , permutation feature importance

References

• Dralle H, Machens A, Basa J, Fatourechi S, Hay ID, et al. Follicular cell-derived thyroid cancer. Nat Rev Dis Primers. 2015;1:15077.
• Zhao H, Liu CH, Cao Y, Zhang LY, Zhao Y, Liu YW, et al. Survival prognostic factors for differentiated thyroid cancer patients, with pulmonary metastases: A systematic review and meta-analysis. Front Oncol. 2022;12:990154.
• Na’ara S, Amit M, Fridman E, Gil Z. Contemporary management of recurrent nodal disease in differentiated thyroid carcinoma. Rambam Maimonides Med J. 2016;7(1):e0006.
• Clark E, Price S, Lucena T, Haberlein B, Wahbeh A, Seetan R. Predictive analytics for thyroid cancer recurrence: a machine learning approach. Knowledge. 2024;4(4):557-570.
• Schindele A, Krebold A, Heiß U, Nimptsch K, Pfaehler E, Berr C, et al. Interpretable machine learning for thyroid cancer recurrence prediction: Leveraging XGBoost and SHAP analysis. Eur J Radiol. 2025;186:112049.
• Medas F, Canu GL, Boi F, Lai ML, Erdas E, Calò PG. Predictive factors of recurrence in patients with differentiated thyroid carcinoma: A retrospective analysis on 579 patients. Cancers (Basel). 2019;11(9):1230.
• Adlung L, Cohen Y, Mor U, Elinav E. Machine learning in clinical decision making. Med. 2021;2(6):642-665. Taha K. Machine learning in biomedical and health big data: a comprehensive survey with empirical and experimental insights. J Big Data. 2025;12(1):61.
• Wang H, Zhang C, Li Q, Tian T, Huang R, Qiu J, et al. Development and validation of prediction models for papillary thyroid cancer structural recurrence using machine learning approaches. BMC Cancer. 2024;24(1):427.
• Borzooei S, Briganti G, Golparian M, Lechien JR, Tarokhian A. Machine learning for risk stratification of thyroid cancer patients: a 15-year cohort study. Eur Arch Otorhinolaryngol. 2024;281(4):2095-2104.
• Setiawan KE. Predicting recurrence in differentiated thyroid cancer: a comparative analysis of various machine learning models including ensemble methods with chi-squared feature selection. Commun Math Biol Neurosci. 2024;2024:55.
• Yasar S. Determination of possible biomarkers for predicting well-differentiated thyroid cancer recurrence by different ensemble machine learning methods. Middle Black Sea J Health Sci. 2024;10(3):255-265.
• Frasca M, La Torre D, Pravettoni G, Cutica I. Explainable and interpretable artificial intelligence in medicine: a systematic bibliometric review. Discov Artif Intell. 2024;4(1):15.
• Amann J, Blasimme A, Vayena E, Frey D, Madai VI. Explainability for artificial intelligence in healthcare: a multidisciplinary perspective. BMC Med Inform Decis Mak. 2020;20:1-9.
• Minh D, Wang HX, Li YF, Nguyen TN. Explainable artificial intelligence: a comprehensive review. Artif Intell Rev. 2020;1-66.
• UCI Machine Learning Repository. Differentiated thyroid cancer recurrence dataset. Available from: https://archive.ics.uci.edu/dataset/915/differentiated+thyroid+cancer+recurrence [Accessed 2024 Jun 4].
• Yin Y, Jang-Jaccard J, Xu W, Singh A, Zhu J, Sabrina F, et al. IGRF-RFE: a hybrid feature selection method for MLP-based network intrusion detection on UNSW-NB15 dataset. J Big Data. 2023;10(1):15.
• Mohandoss DP, Shi Y, Suo K. Outlier prediction using random forest classifier. In: 2021 IEEE 11th Annual Computing and Communication Workshop and Conference (CCWC); January 2021. p. 27-33. IEEE.
• Zhang C, Shao X, Li D. Knowledge-based support vector classification based on C-SVC. Procedia Comput Sci. 2013;17:1083-1090.
• Schapire RE. The boosting approach to machine learning: An overview. In: Denison DD, Hansen MH, Holmes CC, Mallick B, Yu B, eds. Nonlinear Estimation and Classification. Lecture Notes in Statistics, vol 171. Springer; 2003. p. 203-225.
• LaValley MP. Logistic regression. Circulation. 2008;117(18):2395-2399.
• Baghdadi NA, Farghaly Abdelaliem SM, Malki A, Gad I, Ewis A, Atlam E. Advanced machine learning techniques for cardiovascular disease early detection and diagnosis. J Big Data. 2023;10(1):144.
• Band SS, Yarahmadi A, Hsu CC, Biyari M, Sookhak M, Ameri R, et al. Application of explainable artificial intelligence in medical health: a systematic review of interpretability methods. Informatics Med Unlocked. 2023;40:101286.
• Altmann A, Toloşi L, Sander O, Lengauer T. Permutation importance: a corrected feature importance measure. Bioinformatics. 2010;26(10):1340-1347.
• Mosca E, Szigeti F, Tragianni S, Gallagher D, Groh G. SHAP-based explanation methods: a review for NLP interpretability. In: Proceedings of the 29th International Conference on Computational Linguistics; October 2022. p. 4593-4603.
• Cutler A, Cutler DR, Stevens JR. Random forests. In: Ensemble Machine Learning: Methods and Applications. Springer; 2012. p. 157-175.
• Kök I, Okay FY, Muyanlı O, Ozdemir S. Explainable artificial intelligence (XAI) for Internet of Things: a survey. IEEE Internet Things J. 2023;10(16):14764-14779.
• Ruben R, Pavithran PV, Menon VU, Nair V, Kumar H. Performance of ATA risk stratification systems, response to therapy, and outcome in an Indian cohort of differentiated thyroid carcinoma patients: a retrospective study. Eur Thyroid J. 2019;8(6):312-318.
• Tuttle RM, Tala H, Shah J, Leboeuf R, Ghossein R, Gonen M, et al. Estimating risk of recurrence in differentiated thyroid cancer after total thyroidectomy and radioactive iodine remnant ablation: using response to therapy variables to modify the initial risk estimates predicted by the new American Thyroid Association staging system. Thyroid. 2010;20(12):1341-1349.
• Park S, Kim WG, Song E, Oh HS, Kim M, Kwon H, et al. Dynamic risk stratification for predicting recurrence in patients with differentiated thyroid cancer treated without radioactive iodine remnant ablation therapy. Thyroid. 2017;27(4):524-530.
• Ito Y, Miyauchi A, Kihara M, Fukushima M, Higashiyama T, Miya A. Overall survival of papillary thyroid carcinoma patients: a single-institution long-term follow-up of 5897 patients. World J Surg. 2018;42:615-622.
• Altay FP, Cicek O, Demirkan E, Taşkaldiran I, Bozkus Y, Turhan Iyidir O, et al. Evaluation of prognosis and risk factors of differentiated thyroid cancer in a geriatric population. Turk J Geriatr. 2023;26:118-123.