Artificial Intelligence Fights Thyroid Cancer: Accurate Prediction of Thyroid Cancer Recurrence Risk through Machine Learning Models
DOI:
https://doi.org/10.59994/ajamts.2025.1.10Keywords:
Artificial Intelligence, Thyroid Cancer, Cancer Recurrence Prediction, Machine Learning, Predictive ModelingAbstract
The aim of this study is to determine the possible use of machine learning models in improving predictive accuracy of recurrence of thyroid cancer, particularly Differentiated Thyroid Cancer (DTC). A cohort of 383 patients who underwent radioactive iodine treatment was studied with six machine learning models, i.e., Random Forest, Decision Tree, and K-Nearest Neighbors. Experiments showed that Random Forest was the best-performing model, which had 98.7% test set accuracy, showing that it had strong predictive power with accuracy and reliability. Data rebalancing solved data imbalance issues, as well as the utilization of methods for hyperparameter tuning. Feature selection and the use of laboratory and clinical data in improving performance were also a high priority in the research. But more research will be required to enhance clinical transparency and develop models on more heterogeneous data, incorporating genetic and biochemical biomarkers.
References
[1] D. S. Ross et al., “2016 American Thyroid Association guidelines for diagnosis and management of hyperthyroidism and other causes of thyrotoxicosis,” Thyroid, vol. 26, no. 10, pp. 1343–1421, 2016.
[2] F. Bini et al., “Artificial intelligence in thyroid field—a comprehensive review,” Cancers, vol. 13, no. 19, p. 4740, 2021.
[3] M. Zeer, M. R. A. Sara, A. Sbeih, and K. Sabarna, “Predicting the Risk of Myocardial Infarction (MI) using Machine Learning (ML),” Ahliya J. Allied Medico-Technol. Sci., vol. 1, no. 1, pp. 26–29, 2024.
[4] I. Imtiaz et al., “The Future of Differentiated Thyroid Cancer Recurrence Prediction Using a Machine Learning Framework Advancements, Challenges, and Prospects,” in 2024 International Conference on Emerging Trends in Networks and Computer Communications (ETNCC), IEEE, 2024, pp. 518–525.
[5] E. Clark, S. Price, T. Lucena, B. Haberlein, A. Wahbeh, and R. Seetan, “Predictive analytics for thyroid cancer recurrence: A machine learning approach,” Knowledge, vol. 4, no. 4, pp. 557–570, 2024.
[6] S. Chattopadhyay, “Towards Predicting Recurrence Risk of Differentiated Thyroid Cancer with a Hybrid Machine Learning Model,” Medinformatics, 2024.
[7] A. Schindele et al., “Interpretable machine learning for thyroid cancer recurrence predicton: leveraging XGBoost and SHAP analysis,” Eur. J. Radiol., vol. 186, p. 112049, 2025.
[8] F. Firat Atay et al., “A hybrid machine learning model combining association rule mining and classification algorithms to predict differentiated thyroid cancer recurrence,” Front. Med., vol. 11, p. 1461372, 2024.
[9] R. T. Kloos, “Thyroid cancer recurrence in patients clinically free of disease with undetectable or very low serum thyroglobulin values,” J. Clin. Endocrinol. Metab., vol. 95, no. 12, pp. 5241–5248, 2010.
