Machine learning-based prediction model for Fear of progression in thyroid cancer survivors.

OBJECTIVE

This study aimed to develop and compare machine learning prediction models for Fear of Progression (FOP) in thyroid cancer survivors, providing an effective tool for early clinical identification of high-risk populations.

METHODS

A cross-sectional study design was adopted, enrolling 356 thyroid cancer patients, among whom 33.5% exhibited clinically significant FOP symptoms. Predictive variables were systematically collected, including clinical characteristics, treatment regimens, and psychosocial factors. Five machine learning algorithms (logistic regression, elastic net, support vector machine, random forest, and XGBoost) were employed to construct prediction models. Model performance was evaluated using metrics such as the area under the receiver operating characteristic curve (AUC), accuracy, and sensitivity.

RESULTS

Univariate analysis identified several factors significantly associated with FOP, including advanced tumor stage (III–IV), total thyroidectomy, adjuvant radioactive iodine therapy, lower psychological resilience, and inadequate social support. (P < 0.05). Comparative analysis of machine learning models demonstrated that the XGBoost model achieved the highest predictive performance (AUC = 0.84, accuracy = 85.14%), followed by the random forest model (AUC = 0.83, sensitivity = 77%), while the remaining models exhibited relatively lower predictive efficacy.

CONCLUSION

The XGBoost-based prediction model demonstrated superior performance in assessing FOP risk among thyroid cancer patients, serving as an effective clinical screening tool for high-risk populations and providing a scientific basis for early psychological intervention.