Premium
Positional Calibration of an Ultrasound Image‐Guided Robotic Breast Biopsy System
Author(s) -
Nelson Thomas R.,
Tran Amy,
Fakourfar Hourieh,
Nebeker Jakob
Publication year - 2012
Publication title -
journal of ultrasound in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.574
H-Index - 91
eISSN - 1550-9613
pISSN - 0278-4297
DOI - 10.7863/jum.2012.31.3.351
Subject(s) - medicine , breast biopsy , biopsy , sampling (signal processing) , breast cancer , robotic arm , ultrasound , artificial intelligence , radiology , computer vision , image quality , computer science , cancer , mammography , image (mathematics) , filter (signal processing)
Objectives Precision biopsy of small lesions is essential in providing high‐quality patient diagnosis and management. Localization depends on high‐quality imaging. We have developed a dedicated, fully automatic volume breast ultrasound (US) imaging system for early breast cancer detection. This work focuses on development of an image‐guided robotic biopsy system that is integrated with the volume breast US system for performing minimally invasive breast biopsies. The objective of this work was to assess the positional accuracy of the robotic system for breast biopsy. Methods We have adapted a compact robotic arm for performing breast biopsy. The arm incorporates a force torque sensor and is modified to accommodate breast biopsy sampling needles mounted on the robot end effector. Volume breast US images are used as input to a targeting algorithm that provides the physician with control of biopsy device guidance and trajectory optimization. In this work, the positional accuracy was evaluated using (1) a light‐emitting diode (LED) mounted on the end effector and (2) a LED mounted on the end of a biopsy needle, each of which was imaged for each robot controller position as part of mapping the positional accuracy throughout a volume that would contain the breast. We measured the error in each location and the cumulative error. Results Robotic device performance over the volume provided mean accuracy ± SD of 0.76 ± 0.13 mm (end effector) and 0.55 ± 0.13 mm (needle sample location), sufficient for a targeting accuracy within ±1 mm, which is suitable for clinical use. Depth positioning error also was small: 0.38 ± 0.03 mm. Reproducibility was excellent with less than 0.5% variation. Conclusions Overall accuracy and reproducibility of the compact robotic device were excellent, well within clinical biopsy performance requirements. Volume breast US data provide high‐quality input to a biopsy sampling algorithm under physician control. Robotic devices may provide more precise device placement, assisting physicians with biopsy procedures.