Suppr超能文献

基于M型基准体模的快速自动超声校准技术:一种二维到三维图像配准的新方法。

M-fiducial phantom based fast automatic ultrasound calibration technique: A novel approach for 2D to 3D image registration.

作者信息

Suthakorn Jackrit, Chatrasingh Maria, Wiratkapun Cholatip, Ongwattanakul Songpol

机构信息

Department of Biomedical Engineering, Center for Biomedical and Robotics Technology (BART LAB), Faculty of Engineering, Mahidol University, Salaya, Thailand.

Department of Radiology, Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand.

出版信息

Heliyon. 2024 Nov 8;10(22):e40290. doi: 10.1016/j.heliyon.2024.e40290. eCollection 2024 Nov 30.

Abstract

This study introduces a novel M-fiducial phantom-based automatic ultrasound (US) calibration technique aimed at improving 2D to 3D image registration in ultrasound-guided procedures. The technique addresses limitations in existing methods by enhancing feature extraction capabilities and improving calibration accuracy. The proposed method utilizes an M-fiducial phantom, an extended version of the standard N-fiducial phantom, designed to facilitate automatic extraction with a linear probe, particularly in small field-of-view scenarios. A closed-form solution incorporating anisotropic scaling enables rapid and precise computation of transformation matrices for ultrasound calibration, and the method was tested across varying ultrasound depths to evaluate performance in point localization, length measurement, and volume estimation. Results show that the automatic extraction method, combined with the anisotropic closed-form solution, offers greater accuracy and reliability compared to manual methods. While in-plane measurements demonstrated high accuracy, out-plane estimations revealed potential limitations, especially with wire phantoms at increased depths. Despite these challenges, the M-fiducial phantom-based technique reduces operator workload and enhances ultrasound calibration precision, making it a promising approach for high-precision ultrasound-guided interventions. Future work should focus on refining the technique and validating its efficacy in clinical applications.

摘要

本研究介绍了一种基于新型M型基准体模的自动超声(US)校准技术,旨在改善超声引导手术中的二维到三维图像配准。该技术通过增强特征提取能力和提高校准精度来解决现有方法的局限性。所提出的方法利用了M型基准体模,它是标准N型基准体模的扩展版本,设计用于使用线性探头进行自动提取,特别是在小视野场景中。一种包含各向异性缩放的闭式解能够快速精确地计算超声校准的变换矩阵,并且该方法在不同的超声深度下进行了测试,以评估其在点定位、长度测量和体积估计方面的性能。结果表明,与手动方法相比,自动提取方法与各向异性闭式解相结合具有更高的准确性和可靠性。虽然平面内测量显示出高精度,但平面外估计揭示了潜在的局限性,特别是在深度增加时使用线阵体模的情况下。尽管存在这些挑战,基于M型基准体模的技术减少了操作员的工作量并提高了超声校准精度,使其成为高精度超声引导介入的一种有前途的方法。未来的工作应集中在改进该技术并在临床应用中验证其有效性。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/beb9/11605415/eb772be103ba/ga1.jpg

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验