Clinical researchers at Nagoya City University in Japan have tested an AI-powered surgical robot in its ability to assist with percutaneous nephrolithotomy, which is a minimally invasive procedure to remove large kidney stones. The technique involves accessing the kidney through the skin, and typically requires a highly experienced surgeon. The robot, called the Automated Needle Targeting with X-ray (ANT-X) was developed by NDR Medical Technology, a medical startup based in Singapore. The company reports that the robot can assist with needle placement and can calculate needle targeting in a few seconds, using just one X-ray image. In the trial, the robot made the procedure much easier.
Percutaneous nephrolithotomy (PCNL) involves accessing the kidney through the skin of the back. The procedure is challenging and achieving such renal access incorrectly can lead to serious complications, including bowel perforation, serious bleeding, and sepsis. Consequently, surgeons must undergo years of training before performing the procedure routinely.
The ANT-X robot is designed to assist with the procedure, and potentially make it safer for a less experienced surgeon to perform. The system is intended to be used in conjunction with C-arm fluoroscopy and the robot uses AI to analyze a single X-ray image and then automate needle alignment. In theory, this frees the surgeon up to focus on the depth of insertion.
This latest study compared the robot with conventional ultrasound-guided PCNL in a group of patients at Nagoya City University Hospital. A single, novice surgeon performed the robot-assisted renal access procedures, whereas multiple experienced surgeons performed the conventional renal access procedures.
Achieving renal access with just one needle puncture occurred 34% and 50% of the time in the conventional and robot-assisted groups, respectively. Moreover, the average number of required needle punctures was less in the robot group compared to the conventional group (1.82 times vs 2.51 times). Strikingly, in 14.3% of the procedures in the conventional group, the surgeon could not achieve renal access, and a different surgeon had to help. This did not happen in the robot-assisted group.
The results illustrate the utility and safety of AI-powered robots in performing fiddly surgical tasks. “The ANT-X simplifies a complex procedure, like the PCNL, making it easier for more doctors to perform it and help more patients in the process,” said Kazumi Taguchi, a researcher involved in the study. “Being an AI-powered robotic technology, this technique may pave the way for automating similar interventional surgeries that could shorten the procedure time, relieve the burden off of senior doctors, and perhaps reduce the occurrence of complications.”