TY - JOUR
T1 - Feasibility of Augmented Reality–Guided Transjugular Intrahepatic Portosystemic Shunt
AU - Yang, Jian
AU - Zhu, Jianjun
AU - Sze, Daniel Y.
AU - Cui, Li
AU - Li, Xiaohui
AU - Bai, Yanhua
AU - Ai, Danni
AU - Fan, Jingfan
AU - Song, Hong
AU - Duan, Feng
N1 - Publisher Copyright:
© 2020 SIR
PY - 2020/12
Y1 - 2020/12
N2 - Purpose: To investigate an augmented reality (AR)–guided endovascular puncture to facilitate successful transjugular intrahepatic portosystemic shunt (TIPS). Materials and Methods: An AR navigation system for TIPS was designed. Three-dimensional (3D) liver models including portal and hepatic vein anatomy were extracted from preoperative CT images. The 3D models, intraoperative subjects, and electromagnetic tracking information of the puncture needles were integrated through the system calibration. In the AR head-mounted display, the 3D models were overlaid on the subjects, which was a liver phantom in the first phase and live beagle dogs in the second phase. One life-size liver phantom and 9 beagle dogs were used in the experiments. Imaging after puncture was performed to validate whether the needle tip accessed the target hepatic vein successfully. Results: Endovascular punctures of the portal vein of the liver phantom were repeated 30 times under the guidance of the AR system, and the puncture needle successfully accessed the target vein during each attempt. In the experiments of live canine subjects, the punctures were successful in 2 attempts in 7 beagle dogs and in 1 attempt in the remaining 2 dogs. The puncture time of needle from hepatic vein to portal vein was 5–10 s in the phantom experiments and 10–30 s in the canine experiments. Conclusions: The feasibility of AR-based navigation facilitating accurate and successful portal vein access in preclinical models of TIPS was validated.
AB - Purpose: To investigate an augmented reality (AR)–guided endovascular puncture to facilitate successful transjugular intrahepatic portosystemic shunt (TIPS). Materials and Methods: An AR navigation system for TIPS was designed. Three-dimensional (3D) liver models including portal and hepatic vein anatomy were extracted from preoperative CT images. The 3D models, intraoperative subjects, and electromagnetic tracking information of the puncture needles were integrated through the system calibration. In the AR head-mounted display, the 3D models were overlaid on the subjects, which was a liver phantom in the first phase and live beagle dogs in the second phase. One life-size liver phantom and 9 beagle dogs were used in the experiments. Imaging after puncture was performed to validate whether the needle tip accessed the target hepatic vein successfully. Results: Endovascular punctures of the portal vein of the liver phantom were repeated 30 times under the guidance of the AR system, and the puncture needle successfully accessed the target vein during each attempt. In the experiments of live canine subjects, the punctures were successful in 2 attempts in 7 beagle dogs and in 1 attempt in the remaining 2 dogs. The puncture time of needle from hepatic vein to portal vein was 5–10 s in the phantom experiments and 10–30 s in the canine experiments. Conclusions: The feasibility of AR-based navigation facilitating accurate and successful portal vein access in preclinical models of TIPS was validated.
UR - http://www.scopus.com/inward/record.url?scp=85097037957&partnerID=8YFLogxK
U2 - 10.1016/j.jvir.2020.07.025
DO - 10.1016/j.jvir.2020.07.025
M3 - Article
C2 - 33261744
AN - SCOPUS:85097037957
SN - 1051-0443
VL - 31
SP - 2098
EP - 2103
JO - Journal of Vascular and Interventional Radiology
JF - Journal of Vascular and Interventional Radiology
IS - 12
ER -