Image-Guided Robotic Interventions for Core Needle Biopsy

LIM-DISSERTATION-2019.pdf (15.34 MB)
LIM-DISSERTATION-2019.docx (30.75 MB)
Embargo until
Journal Title
Journal ISSN
Volume Title
Johns Hopkins University
Image-guided core needle biopsy (CNB) is a common procedure to diagnose cancer in patients with clinical symptoms and/or cancer suspicious regions (CSR) of imaging abnormality. While image-guided CNB can be performed manually, the quality control of the procedure is often subjective, outcomes are variable depending on the training and skills of the physicians, and significant challenges exist in navigating the biopsy needle accurately under image-guidance. Robotic assistance for needle guidance has the potential to increase needle targeting accuracy in CNB. This work presents two types of robotic systems for image-guided CNB. The first system is a magnetic resonance imaging (MRI)-guided robotic system, which was developed to assist pediatricians in performing bone biopsy. A bone biopsy is frequently needed if a CSR is observed on MRI. Currently, the biopsy can only be performed in open surgery, under X-ray, or computed tomography (CT) guidance. These involve either general anesthesia or exposure to ionizing radiation, respectively, which are especially concerning in children. Most importantly, neither method can use direct MRI as feedback to guide the biopsy. Bone marrow lesions are difficult to visualize during surgery or with X-ray and CT imaging, increasing the possibility of missed sampling and inaccurate diagnosis. Instead, direct MRI-guided targeted robotic biopsy may be performed in the same session with the MRI diagnosis, and allow direct confirmation of needle sampling the CSR. This may also reduce trauma and eliminate radiation exposure in children. Comprehensive validation tests including bench, mockup tests, and human cadaveric studies are presented, showing the feasibility of the system. The second system is a transrectal ultrasound (TRUS)-guided robotic system, which was developed to assist urologists in performing prostate biopsy. Current challenges and deficiencies of manual biopsy are: the difficulty navigating the needle under 2D ultrasound guidance; the deformations of the prostate by free-hand probe manipulation during biopsy; and the use of the same biopsy plan for all patients. Instead, the TRUS robot enables accurate needle targeting under 3D ultrasound guidance, with minimized prostate deformations, and personalized for each patient to maximize the likelihood of detecting cancer. Extensive pre-clinical validation tests and clinical trial results are presented.
Medical Robots, Image-Guided Intervention, Core Needle Biopsy, Prostate Biopsy, Bone Biopsy