TY - JOUR
T1 - Review of robot-assisted medical ultrasound imaging systems
T2 - Technology and clinical applications
AU - Huang, Qinghua
AU - Zhou, Jiakang
AU - Li, Zhi Jun
N1 - Publisher Copyright:
© 2023 Elsevier B.V.
PY - 2023/11/28
Y1 - 2023/11/28
N2 - Robot-assisted Medical Ultrasound Imaging Systems (RMUIS) leverage the accuracy and stability of robotic motion and offer the potential to standardize medical ultrasound imaging. RMUIS expand the scope of medical ultrasound applications and promote ultrasound in a wider range of application scenarios and imaging modalities. Compared to free-hand ultrasound, RMUIS can interpret image information more quantitatively and acquire stable images over a long period of time. This not only reduces the operator's workload, but also provides better quantitative results in elastography and 3D imaging. In addition, RMUIS can be better combined with interventional treatment, such as percutaneous biopsy, radiofrequency ablation, thrombolytic therapy, etc. However, autonomous RMUIS and other applications are still in the research phase. Insufficient interactivity, effectiveness, and safety prevent RMUIS from large-scale clinical application. There is an urgent need for intelligent learning to improve the capabilities of RMUIS. In addition, RMUIS is a multidisciplinary system of software and hardware. Hardware modifications and upgrades are also important. In this paper, we review the state of the art of RMUIS and summarize the key technical topics. In each topic, we present a variety of system composition options according to the classification of hardware. A variety of software and algorithms based on different hardware are also presented. We then present the applications of RMUIS in healthcare, showing how the capabilities of robots can be used to improve healthcare. Finally, we summarize the current challenges of RMUIS and propose future directions to explore more diverse applications and promote the practical application of RMUIS in medical scenarios.
AB - Robot-assisted Medical Ultrasound Imaging Systems (RMUIS) leverage the accuracy and stability of robotic motion and offer the potential to standardize medical ultrasound imaging. RMUIS expand the scope of medical ultrasound applications and promote ultrasound in a wider range of application scenarios and imaging modalities. Compared to free-hand ultrasound, RMUIS can interpret image information more quantitatively and acquire stable images over a long period of time. This not only reduces the operator's workload, but also provides better quantitative results in elastography and 3D imaging. In addition, RMUIS can be better combined with interventional treatment, such as percutaneous biopsy, radiofrequency ablation, thrombolytic therapy, etc. However, autonomous RMUIS and other applications are still in the research phase. Insufficient interactivity, effectiveness, and safety prevent RMUIS from large-scale clinical application. There is an urgent need for intelligent learning to improve the capabilities of RMUIS. In addition, RMUIS is a multidisciplinary system of software and hardware. Hardware modifications and upgrades are also important. In this paper, we review the state of the art of RMUIS and summarize the key technical topics. In each topic, we present a variety of system composition options according to the classification of hardware. A variety of software and algorithms based on different hardware are also presented. We then present the applications of RMUIS in healthcare, showing how the capabilities of robots can be used to improve healthcare. Finally, we summarize the current challenges of RMUIS and propose future directions to explore more diverse applications and promote the practical application of RMUIS in medical scenarios.
KW - Image processing
KW - Pattern recognition
KW - Robot-assisted ultrasound system
KW - Robotics
UR - http://www.scopus.com/inward/record.url?scp=85172911523&partnerID=8YFLogxK
U2 - 10.1016/j.neucom.2023.126790
DO - 10.1016/j.neucom.2023.126790
M3 - 短篇评述
AN - SCOPUS:85172911523
SN - 0925-2312
VL - 559
JO - Neurocomputing
JF - Neurocomputing
M1 - 126790
ER -