Robot-assisted CT-guided Biopsy with an Artificial Intelligence-based Needle-path Generator: A Phantom Study

Wednesday, March 8, 2023

4:21 PM – 4:30 PM

Location: Phoenix Convention Center, 221AB

Presenting Author(s)

Tatsuya Kawai, MD, PhD

Assistant professor
Nagoya City University

Disclosure(s): NDR Medical Technology Pte. Ltd.: Contracted Research (Ongoing)

Author/Co-author(s)

Masashi Shimohira, MD, PhD

Associate professor
Department of Radiology, Nagoya City University Graduate School of Medical Sciences
Kn

Keita nakayama, MD, PhD

Assistant professor
Department of Radiology, Nagoya City University Graduate School of Medical Sciences
TS

Takafumi Sato, MD

Fellow
Department of Radiology, Nagoya City University Hospital
KO

Kengo Ohta, MD, PhD

Assistant professor
Department of Radiology, Nagoya City University Graduate School of Medical Sciences
KS

Kazushi Suzuki, MD, PhD

Assistant professor
Department of Radiology, Nagoya City University Graduate School of Medical Sciences
YS

Yusuke Sawada, MD, PhD

Assistant professor
Dept of Radiology, Nagoya City Univ Grad School of
KW

Ka Wei Ng, PhD

Chief technology officer
NDR Medical Technology Pte. Ltd.
SH

Siang Huei Leong, PhD

Chief scientific officer
NDR Medical Technology Pte. Ltd.
AH

Akio Hiwatashi, MD, PhD

Professor
Department of Radiology, Nagoya City University Graduate School of Medical Sciences

Disclosure(s):

Tatsuya Kawai, MD, PhD: NDR Medical Technology Pte. Ltd.: Contracted Research (Ongoing)

Masashi Shimohira, MD, PhD: No financial relationships to disclose

Purpose: To investigate the feasibility of an automated lesion-targeting robotic system in the CT-guided percutaneous procedure. The accuracy, efficiency, and dependency on the operator's proficiency were compared to those in the manual needle insertion procedure.

Materials and Methods: A robotic system, the Automated Needle Targeting device for CT (ANT-C), was used for lesion detection, trajectory generation, and needle-holder positioning. Eight nodules with a varied diameter (ranging 8 to 12 mm) in an abdomen phantom containing imitated vertebrae and ribs were used for the targets. A preoperative CT scan that integrates the coordinates of the nodules and fiducial markers on the robotic arm into a single space was used for planning. The phantom was scanned in a supine or prone orientation. The built-in analyzing software located all the target points and generated the needle trajectories, then the robotic arm equipped with a needle holder was driven to the position through which nodules were targeted for needle insertion. Four experienced interventional radiologists (IR) and four novice doctors (non-IR) inserted 18-G biopsy needles using the ANT-C robotic system or CT-fluoroscopy-guided freehand maneuver. Post-needle insertion CT scans were acquired to assess accuracy, measuring 3-dimensional (3D), lateral, and depth deviations.

Results: A total of 96 needle insertions by robotic assistance and 32 CT-fluoroscopy-guided freehand needle insertions were carried out by IR and non-IR groups. The 3D deviation and depth deviation of the robot-assisted insertion, including all in-plane and oblique out-of-plane trajectories, were significantly smaller than that of the freehand insertion (3.8 ± 1.3 vs. 4.7 ± 1.6 mm; mean ± standard deviation, P < 0.01, and 1.8 ± 1.2 vs. 2.6 ± 1.8 mm, P = 0.023, respectively), where there was no significant difference in the lateral deviation (3.0 ± 1.5 vs. 3.5 ± 1.5 mm, P = 0.13). The insertion time was significantly shorter when using robotic assistance than freehand insertion (17.3 ± 7.8 vs. 78.6 ± 38.1 sec, P < 0.0001). There were no significant differences in the 3D deviation and insertion time between IR and non-IR groups using the robotic assistance (3.6 ± 1.5 vs. 3.9 ± 1.2 mm, P = 0.25, and 15.8 ± 8.4 vs. 18.8 ± 6.9 sec, P = 0.065, respectively).

Conclusion: The ANT-C system was successfully used for a needle targeting procedure in a phantom trial and was shown to provide greater accuracy, including for oblique out-of-plane needle insertions and shorter time during needle puncture.