相关实验视频

Updated: Sep 9, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

基于人工智能的TIRADS算法的诊断性能与已建立的甲状腺结节分类系统的比较

Abdilkadir Bozkuş¹, Yeliz Başar¹, Koray Güven²

¹Acıbadem Maslak Hospital, Clinic of Radiology, İstanbul, Türkiye.

Diagnostic and interventional radiology (Ankara, Turkey)

|September 1, 2025

概括

与其他系统相比,基于人工智能的甲状腺成像报告和数据系统 (AI-TIRADS) 在描述甲状腺结节方面具有更高的特异性. 这种AI-TIRADS可以减少不必要的活检,并改善患者对甲状腺结节的治疗.

科学领域:

放射学
癌症学
人工智能

背景情况:

甲状腺结节很常见,准确的风险分层对于适当的管理至关重要.
有各种甲状腺成像报告和数据系统 (TIRADS) 用于标准化结节评估.
像AI-TIRADS这样的新系统的诊断性能需要彻底评估.

研究的目的:

为了比较多个TIRADS分类的诊断性能,包括AI-TIRADS,用于甲状腺结节.
评估AI-TIRADS在提高特异性和减少不必要的活检方面的有效性.
评估可疑超声波特征对恶性瘤风险的影响.

主要方法:

从1,139名细胞病理诊断患者的1,322个甲状腺结节的回顾性分析.
三位独立放射科医生使用ACR-TIRADS,ATA-TIRADS,EU-TIRADS,K-TIRADS和AI-TIRADS对结节的评估.
诊断性能指标的评估,包括灵敏度,特异性,PPV和NPV,使用McNemar测试进行比较.

主要成果:

与ACR-TIRADS (44. 6%),ATA-TIRADS (39. 3%),EU-TIRADS (40. 1%) 和K-TIRADS (40. 1%) 相比,AI-TIRADS的特异性更高 (53. 6%).
在不影响诊断准确性的情况下,AI- TIRADS 保持了高灵敏度.
多个可疑的超声波特征的存在显著增加了恶性瘤的风险.

结论:

通过提高诊断特异性,AI-TIRADS显著增强甲状腺结节风险分层.
AI- TIRADS可能会减少不必要的甲状腺活检,并优化患者的治疗.
这项研究强调了AI-TIRADS在提高甲状腺结节评估医疗资源利用效率方面的潜力.

关键词:

甲状腺成像报告和数据系统人工智能甲状腺癌甲状腺结节超声波检查

更多相关视频

Author Spotlight: Integrating Ultrasound Imaging with Biochemical Markers for Thyroid Disease Diagnosis

Author Spotlight: Integrating Ultrasound Imaging with Biochemical Markers for Thyroid Disease Diagnosis

Published on: February 9, 2024

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

相关概念视频

Receiver Operating Characteristic Plot

Receiver Operating Characteristic Plot

A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...