Abstract Background/Introduction Despite best ergonomic practices, the physical strain of echocardiography is known to result in injuries in sonographers, impacting both their health and patient care. Purpose We developed a robotic-assisted echocardiography system that enables a sonographer to perform echocardiography without the need to hold an ultrasound probe. We conducted a pilot study of the system within a clinical environment on 24 patients. Methods A robotic arm (UR10, Universal Robots, Odense, Denmark) was modified to hold an ultrasound probe, which was connected to a commercially available echocardiography scanner. The sonographer controls the probe's position on the patient's chest using a hand-held remote stick controller (figure 1). The robotic arm is set to deliver no more than 5 newtons of pressure to the patient's chest wall. Two-dimensional (2D) and three-dimensional (3D) images were acquired from the parasternal and apical windows. Acquired images were compared to routine clinical echocardiography obtained within the prior 24 hours and assessed for both image quality and ejection fraction measurements. Image quality was assessed on a scale of 0 to 3, where zero indicated unusable and 3 indicated good quality. Results 19 males and five females were scanned. The mean age of the patients was 62.5 years, with a standard deviation (SD) of 10.1 years. Indications for echocardiography included an acute ischemic event in 17, arrhythmia in 3 and heart failure in 4. In two individuals, apical views could not be obtained; parasternal views were obtained in all. No statistical difference was noted in image quality between images acquired by either robotic-assisted echocardiography or traditional echocardiography. For apical images, a mean image quality score of 2.7 (SD 0.38) was noted for robotically acquired images, compared to 2.8 (SD 0.42) for traditional echocardiography. From the parasternal view, the mean image score was 2.83 (SD 0.4) for robot-acquired images compared to 2.83 (SD 0.7) for routine clinical echocardiography. There was also no statistical difference noted in EF between either traditional echocardiography or robotic echocardiography. The mean EF was 42.6% (SD 11.0) for traditional echocardiography and 41.9% (SD 9.5) for robotic-assisted echocardiography. With the low tension delivered by the robotic arm, none of the patients reported discomfort during image acquisition. Conclusion Our pilot clinical study demonstrates that robotic-assisted echocardiography can be performed with no detriment to image quality or ejection fraction (EF) measures compared to traditional echocardiography.
Windram et al. (Thu,) studied this question.