A low-cost high-fidelity ultrasound simulator with the inertial tracking of the probe pose. (February 2017)

• Record Type: Journal Article
• Authors: Farsoni, Saverio; Bonfè, Marcello; Astolfi, Luca
• Is Part Of: Control engineering practice. Volume 59(2017)
• Page Start: 183
• Page End: 193
• Publication Date: 2017-02
• Subjects: Inertial tracking -- Ultrasound simulator -- Sensor fusion -- Medical training -- Real-time algorithms; Automatic control -- Periodicals; 629.89
• DOI: 10.1016/j.conengprac.2016.11.002
• Languages: English
• ISSNs: 0967-0661

Abstract

Abstract: The authors developed a versatile ultrasound simulator. The proposed system achieves the main features of a high-fidelity device exploiting low-cost rapid prototyping hardware. The hand-guided ultrasound simulator probe includes a RFID reader, a 9-DOF inertial sensor unit, consisting of an accelerometer, a magnetometer and a gyroscope, and a microcontroller that performs the real-time data acquisition, the processing and the transmission of the estimated pose information to the visualization system, so that the proper ultrasound view can be generated. Since the probe orientation is the main information involved in the pose reconstruction, this work presents and investigates several tracking methods for the probe orientation, exploiting a sensor fusion technique to filter the noisy measurements coming from inertial sensors. The performances of a Kalman filter, a nonlinear complementary filter and a quaternion-based filter as inertial trackers have been tested by means of a robot manipulator, in terms of readiness, accuracy and stability of the estimated orientation signal. The results show that the nonlinear complementary filter and the quaternion-based filter match all the application requirements (RMSE < 3 deg, variance < 1 deg 2, and settling time < 0.3 s ), and they involve a lower computational time with respect to the Kalman filter. Abstract : Highlights: The paper shows a study of inertial tracking systems for an ultrasound simulator. Three sensor fusion methods are tested by means of a robot manipulator. The influence of magnetic disturbances has been investigated. The overall system provides a low-cost high-fidelity simulator for medical training.