改进的C-TIRADS系统及其诊断价值OA
An improved version of C-TIRADS and its value in thyroid nodule diagnosis
甲状腺影像报告和数据系统(thyroid imaging reporting and data system,TIRADS)是依据甲状腺超声影像表现对甲状腺结节的恶性程度进行标准化分级并提供临床治疗参考的评估体系,在临床上受到了广泛的认同和应用.该文基于经典机器学习算法对中国版甲状腺影像报告和数据系统(C-TIRADS)在可疑特征选取以及特征打分方面进行了优化,并提出了一种改进的 C-TIRADS系统.利用 982个经病理证实的甲状腺结节数据对提出的 TIRADS系统的诊断价值进行了研究.实验结果显示,提出的 TIRADS系统较 C-TIRADS在甲状腺结节的良恶性诊断中表现出更优的诊断性能.
The thyroid imaging reporting and data system(TIRADS)is an evaluation system that standardizes and grades the malignancy of thyroid nodules based on thyroid ultrasound imaging and pro-vides clinical treatment references.It has been widely recognized and applied in clinical practice.This article optimizes the suspicious feature selection and feature scoring of the Chinese version of the thyroid imaging reporting and data system(C-TIRADS)based on classical machine learning algorithms,and proposes an improved version of C-TIRADS.The diagnostic value of the proposed TIRADS was studied using data from 982 pathologically confirmed thyroid nodules.The experimental results show that the proposed TIRADS performs better than C-TIRADS in the diagnosis of benign and malignant thyroid nodules.
陈仁栋;王潇倩;朱浩卓
曲阜师范大学数学科学学院,273165,山东省曲阜市曲阜师范大学数学科学学院,273165,山东省曲阜市曲阜师范大学数学科学学院,273165,山东省曲阜市
信息技术与安全科学
甲状腺结节超声特征自适应提升算法良恶性分级
thyroid noduleultrasound featureadaptive boosting algorithmbenign and malignant clas-sification
《曲阜师范大学学报（自然科学版）》 2026 (1)
73-78,6
国家自然科学基金(12001220).
评论