SingHealth is working to further implement across Singapore a novel AI system that automatically identifies spinal landmarks during anaesthetic procedures. 

Called uSINE, the AI technology was recently implemented in clinical practice at KK Women's and Children's Hospital where it was first tested in 2017.

WHAT IT DOES

Developed in collaboration with the National University of Singapore, uSINE is touted to be the world's first software that uses both ultrasound and AI to automatically pinpoint spinal landmarks during an ultrasound scan. It alerts the anaesthetist in real-time upon identification of the right location and angle of insertion, which is done by a proprietary machine-learning algorithm. 

Based on a clinical study, uSINE was found to achieve a 92% success rate of spinal anaesthesia needle insertion on the first attempt in healthy-weighted women and 82% in obese women. This is compared to the 70% and 43% success rates of the conventional manual method in healthy-weight women and obese women, respectively. 

Its implementation across SingHealth is being planned "as early as this year," while more clinical trials are also being set in Australia, the United States, and Europe.

The AI system has been licensed to local medical device company HiCura Medical for commercialisation.

WHY IT MATTERS

The clinical implementation of uSINE will benefit pregnant women, especially those with "challenging spinal anaesthesia landmarks." According to SingHealth, obese women form the majority of this group. Cases of obesity in pregnancy are growing in Singapore, which currently make up about a tenth of the total pregnant population.

"A doctor typically uses his hands to manually identify the landmark for spinal needle insertion. This requires good knowledge of the anatomy, skills, and experience due to the complexity of the procedure. It becomes more challenging in patients who are obese (body mass index of over 30), who have an abnormal spine, or had previous spine surgery," explained KKH Associate Professor Sng Ban Leong. 

THE LARGER TREND

A combination of AI and robotics has been explored for conducting precise needle placement during minimally invasive surgeries. One such example is the ANT-X (automated needle targeting system with x-ray) by Singaporean company NDR Medical Technology. Claimed to be the first of such in the world, the technology facilitates safe and accurate needle puncture to human organs. It automates the calibration and needle alignment from a single x-ray image within seconds. 

ON THE RECORD

"uSINE leverages AI to improve the success rate of achieving precise spinal needle insertion at the first attempt. This improves the quality of anaesthesia and reduces complications such as nerve irritation, blood collection within the tissues in the spine, or in rare cases, neurological injury," Assoc. Prof. Sng shared.