Autonomous Scanning Target Localization for Robotic Lung Ultrasound Imaging

Under the ceaseless global COVID-19 pandemic, lung ultrasound (LUS) is the emerging way for effective diagnosis and severeness evaluation of respiratory diseases. However, close physical contact is unavoidable in conventional clinical ultrasound, increasing the infection risk for health-care workers. Hence, a scanning approach involving minimal physical contact between an operator and a patient is vital to maximize the safety of clinical ultrasound procedures. A robotic ultrasound platform can satisfy this need by remotely manipulating the ultrasound probe with a robotic arm. This paper proposes a robotic LUS system that incorporates the automatic identification and execution of the ultrasound probe placement pose without manual input. An RGB-D camera is utilized to recognize the scanning targets on the patient through a learning-based human pose estimation algorithm and solve for the landing pose to attach the probe vertically to the tissue surface; A position/force controller is designed to handle intraoperative probe pose adjustment for maintaining the contact force. We evaluated the scanning area localization accuracy, motion execution accuracy, and ultrasound image acquisition capability using an upper torso mannequin and a realistic lung ultrasound phantom with healthy and COVID-19-infected lung anatomy. Results demonstrated the overall scanning target localization accuracy of 19.67 ± 4.92 mm and the probe landing pose estimation accuracy of 6.92 ± 2.75 mm in translation, 10.35 ± 2.97 deg in rotation. The contact force-controlled robotic scanning allowed the successful ultrasound image collection, capturing pathological landmarks.

I. Introduction

As the Coronavirus Disease 2019 (COVID-19) continues to overwhelm global medical resources, a cost-effective diagnostic approach capable of monitoring the severeness of infection on COVID-19 patients is highly desirable. Computed tomography (CT) and X-ray are considered the gold-standard diagnostic imaging for lung-related diseases [1]. However, their accessibility is growingly limited due to the overwhelming amount of COVID-19 patients around the globe [2]. Ultrasound (US) imaging, in comparison, is easier to access, more affordable, radiation-free, highly sensitive to pneumonia, and has the real-time capability [3]. Therefore lung ultrasound (LUS) becomes an alternative accessible approach to diagnose COVID-19 and other contagious lung pathology (e.g., in medical equipment restricted places) [4]. An effective LUS scans a wide area of the chest. Several protocols are introduced to standardize the LUS procedure. Bedside LUS in an emergency modified (BLUE) protocol [5] is an accepted standard in which the anterior chest area is divided into a few regions, and the centroid of each region is considered the scanning target. The operator places the US probe perpendicularly on each target with an appropriate amount of force applied, followed by fine-tuning movements to search for pathological features, and all the targets are to be covered sequentially. However, the LUS procedure is highly operator-dependent and requires physical contact between the operator and patient for a substantial amount of time, increasing the operator’s vulnerability. A less contact-intensive LUS procedure could significantly reduce transmission risk when handling patients with infective respiratory diseases but is difficult to achieve with conventional freehand US. The robotic US (RUS), which uses a robot manipulator equipped with an US transducer to perform an US scan, on the other hand, is a feasible solution to address the clinical need by isolating patients from operators [6]. In recent years, RUS has been actively studied to augment human-operated US [7], [8] for various clinical applications, including thyroid [9], [10], spine [11], vascular [12], [13] and joint [14] imaging. To assure the safety and efficacy of the RUS system, the scanning procedure should be close to the real clinical practice and should avoid excessive pressure being applied on the patient [15].