中国机械工程

• 智能制造 • 上一篇    下一篇

基于改进阻抗控制的自适应曲面乳腺超声扫描方法

沈运1;谢荣理2;赵艳娜3;付庄1;王尧1;张俊2;费健2   

  1. 1. 上海交通大学机械系统与振动国家重点实验室,上海,200240
    2. 上海交通大学医学院附属瑞金医院,上海,200025
    3. 上海市瑞金康复医院,上海,200023
  • 出版日期:2021-01-25 发布日期:2021-02-01
  • 基金资助:
    国家自然科学基金(61973210);
    医工交叉项目(YG2019ZDA17, ZH2018QNB23);
    上海市科学技术委员会项目(17441901000);
    上海市黄浦区科研新项目(HKQ201810)

Adaptive Curved Breast Ultrasound Scanning Method Based on Improved Impedance Control

SHEN Yun1;XIE Rongli2;ZHAO Yanna3;FU Zhuang1;WANG Yao1;ZHANG Jun2;FEI Jian2   

  1. 1. State Key Laboratory of Mechanical System and Vibration, Shanghai Jiao Tong University,Shanghai, 200240
    2. Ruijin Hospital Affiliated to Shanghai Jiao Tong University, Shanghai, 200025
    3. Shanghai Ruijin Rehabilitation Hospital, Shanghai, 200023
  • Online:2021-01-25 Published:2021-02-01

摘要: 针对机器人辅助超声扫描中皮肤曲面贴合的问题,分析了扫描过程中超声探头和皮肤表面的相互作用,并根据其力学规律建立了一种基于改进阻抗控制的扫描算法,以保证探头和皮肤表面之间的法向接触力基本恒定,进而使得扫描过程中超声探头和皮肤表面始终保持法向垂直;根据超声探头的历史轨迹数据,提出了改进阻抗控制的扫描算法来加强曲面贴合的效果;通过乳腺体模和实际人体的测试实验,来验证所提控制算法在曲面自适应贴合问题上的表现性能以及自动扫描方法的完整性。实验结果表明:改进阻抗控制的扫描算法可以取得较佳的恒力控制效果,法向接触力偏差低于±0.2 N,且显著提高了探头转角的跟踪效果,平均转角跟踪误差从4.9°下降到2.2°。乳腺体模和人体活体实验佐证了所提控制算法的有效性和超声扫描的完整性。

关键词: 机器人辅助乳腺超声扫描, 曲面贴合, 阻抗控制, 参数跟踪, 卡尔曼滤波

Abstract: For the problems of skin curved-surface fitting in robot-assisted ultrasonic scanning, the interaction between ultrasonic probe and skin surface during scanning was analyzed. According to the mechanics principle of the interaction, a scanning algorithm via impedance control was first proposed to ensure the consistency of the normal force between the probe and skin surface. Besides, the ultrasonic probe and the skin surface were also kept perpendicular during the scanning processes. Moreover, based on the historical trajectory data of the ultrasonic probe, the scanning algorithm was improved to enhance the surface fitting accuracy. In addition, the experiments on both of breast phantom and actual human body were designed to demonstrate the effectiveness of the designed control algorithm on the surface adaptive fitting problems and the integrity of the proposed scanning method. The experimental results show that the improved impedance control scanning algorithm achieves high performance on constant force control, and the normal force bias was within ±0.2 N. The tracking effectiveness of the probe rotation angles is also significantly improved, and the average tracking errors decrease from 4.9° to 2.2°. Also, the experiments truly confirmed the effectiveness of the control algorithm and the integrity of the ultrasound scanning.

Key words: robot-assisted breast ultrasound scanning, curved-surface fitting, impedance control, parameter tracking, Kalman filtering

中图分类号: