Force-controlled robotic ultrasound elastography enhances diagnostic consistency for thyroid nodules

力控机器人超声弹性成像技术提高了甲状腺结节诊断的一致性

阅读:2

Abstract

OBJECTIVE: To assess whether a force-controlled robotic arm can improve image quality, consistency, and diagnostic accuracy of strain elastography (SE) and shear wave elastography (SWE) in differentiating benign from malignant thyroid nodules. MATERIALS AND METHODS: In this prospective study, 131 thyroid nodules were examined by a junior physician, a senior physician, and a robotic arm with a PID-based force feedback system. Image quality was evaluated using the structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), mean squared error (MSE) and mean opinion score (MOS). Consistency was assessed with the Dice similarity coefficient (DSC) and the intraclass correlation coefficient (ICC) for SWE-derived Emax. Diagnostic performance was analyzed via ROC curves and AUC comparisons. RESULTS: The robotic arm achieved higher image quality (SSIM up to 0.94, PSNR 42.25 dB, MSE 3.06) and MOS scores (SE: 4.1 ± 0.3; SWE: 4.3 ± 0.2) than both human operators (all p < 0.001). It also showed better consistency (DSC up to 0.90; ICC up to 0.94) and diagnostic accuracy (AUC 0.90 for SE, 0.95 for SWE; p < 0.05). CONCLUSION: The force-controlled robotic arm provides standardized, reproducible thyroid elastography with superior quality, consistency, and accuracy compared with manual scanning. CRITICAL RELEVANCE STATEMENT: This study explores robotic-assisted elastography to improve consistency in thyroid nodule assessment. KEY POINTS: The force-controlled robotic arm enhanced ultrasound elastography by generating clearer, more standardized images compared with manual acquisition. The force-controlled robotic arm scanning achieved superior intra-operator reproducibility. Diagnostic performance in differentiating benign from malignant thyroid nodules was significantly improved with robotic elastography, exceeding that of both junior and senior radiologists.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。