Design And Implementation Of Robotic ARM Based Automatic Medical Ultrasound Scanning System For Three-Dimensional Imaging

Medical ultrasound imaging has become one of the most routine and widely used diagnostic methods in modern medicine,because of its advantages of low cost,no radiation,good real-time performance,and rapid diagnosis.As people’s health needs increase,its application is expanding in depth degree and breadth degree.However,the image obtained by traditional two-dimensional(2D)ultrasound comes from a slice of three-dimensional(3D)tissue,which has obvious limitation.The diagnosis results greatly depend on the physician’s scanning skill and experience,and hence may not maintain objectivity.3D ultrasound imaging can display intuitively,the spatial relationship of the tissue anatomical structure is explicit.The tissue and organs can be observed in multi-plane and multi-angle,which can help the doctor accurately diagnose diseases and shorten the scanning time.3D ultrasound imaging can well overcome the limitation of 2D ultrasound.With the development and progress of science and technology,more and more medical robot systems have been developed.Robotic ultrasound systems(RUS)combined the ultrasound examination with the robotic system in medical interventions.Robots can be used to improve the performance of ultrasound systems in data acquisition and manipulation.RUS show great prospects for improving clinical diagnosis and treatment.However,the existing robotic ultrasound system more or less has some limitations,such as a single application scenario and insufficient autonomy,which need to be improved.In the paper,we propose a robotic arm based automatic medical ultrasound scanning system for three-dimensional imaging.The depth camera was adopted to acquire the contour of the tissue surface.The operator could set the scan path using three scan-path planning modes.Then,the system would determine the pose of the robotic arm corresponding to each scan point based on a normal-vector-based method.During the scanning,the system would fine-tune the pose of the probe according to the force feedback between the probe and the tissue surface.Having completed the scanning progress,the system would use the Bezier interpolation method to finish the volume reconstruction and provide 3D visualization.To verify the performance of the proposed system in different tissues,the system conducted scanning experiments with multi-modality breast biopsy and sonographic trainer、thyroid training phantom、thyroid training phantom、lumbar training phantom,and real human forearm.The experimental results show that the system can independently complete the automatic scanning of 3D ultrasound,has good imaging quality and broad clinical application prospects.