Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

An image-based robotic needle guidance system for interventional radiology

: Pätz, Torben; Rothfuss, Andreas

Skalej, M. ; Univ. Magdeburg; Medizinische Hochschule Hannover; Forschungscampus M²OLIE, Mannheim; Forschungscampus STIMULATE, Magdeburg:
3rd Conference on Image-Guided Interventions & Fokus Neuroradiologie 2017. Abstract Book : Recent Progress and Developments. November 6 and 7, 2017 Magdeburg, Germany
Magdeburg, 2017
Conference on Image-Guided Interventions & Fokus Neuroradiologie <3, 2017, Magdeburg>
Fraunhofer IPA ()
Medizintechnik; Bilderkennung; Minimal invasive Chirurgie; Minimal invasive Therapie; Gewebeprobe; Medizinroboter; Radiologie

Intro: We present a pre-clinical robotic guidance system with a dedicated image-based control software to support needle-based interventions in the angio suite. We also present first precision results for needle placement.
Methods: The system consists of a Kuka LBR robot, equipped with a control software with OpenIGTLink interface and an image-based calibration and target planning software based on MeVisLab. The robot can hold either a calibration tool, containing 4 spheres with known geometry for coordinate system alignment of the imaging device and the robot or a needle-guide that prescribes the needle trajectory and supports multi-needle-placement. For needle placement, the robot places the needle guide on the patient skin to determine the needle entry point. The calibration and targeting software provides automatic calibration tool detection, image segmentation, registration, virtual needle placement, and ablation simulation.
For the precision analysis we used a 15cm 16G needle that has been placed at n=13 target positions. For all positions, we compared planned and reached needle path in air and a gelatin phantom.
Results: The robotic guidance system had an overall angular deviation between planned and reached position that is below 1° (range 0.50° to 1.37°), and an overall deviation of the shaft resp. skin entry point of 1.4mm (range 0.69mm to 2.41mm). Reproducibility of the needle path between air and gelatin was far better than overall system precision leaving potential for further system improvements. Time required for robot motion to target, needle insertion, 3D scan, needle removal, and robot motion to rest position was less than two minutes for this technical setup.
Conclusion: The analyzed system has a better angle and shaft precision as reported for comparative systems. It is easy to use and fast. The system has potential to serve as a fully integrated biopsy and therapeutic needle intervention system in head and neck, pulmonary, and abdominal areas.