The Application Value of Deep Learning-Based Nomograms in Benign-Malignant Discrimination of TI-RADS Category 4 Thyroid Nodules

Abstract

Abstract This study employed deep learning techniques to accurately classify TI-RADS category 4 thyroid nodules as either benign or malignant, and developed a nomogram that incorporated relevant clinical factors. A total of 500 patients were included and randomly divided into a training group (350 patients) and a test group (150 patients). The YOLOv3 model was constructed and evaluated using various metrics, achieving an 84% accuracy in classifying TI-RADS category 4 thyroid nodules. Based on the model's predictions, clinical data, and ultrasound data, a nomogram was developed. The nomogram exhibited superior performance in both the training and testing groups. Additionally, the calibration curve demonstrated good agreement between predicted probabilities and actual outcomes. Decision curve analysis further confirmed that the nomogram provided greater net benefits. Ultimately, the YOLOv3 model and nomogram successfully improved the accuracy of distinguishing between benign and malignant TI-RADS category 4 thyroid nodules.