Becker’s Spine Review
An artificial intelligence-powered ultrasound-guided automated spinal landmark identification system improves the accuracy and effectiveness of first-time needle insertion for spinal anesthesia, according to a study conducted in Singapore.
The system uses a machine-learning algorithm to identify spinal landmarks during an ultrasound. An anesthesiologist is alerted in real time when the system identifies the right location and right angle for insertion.
Physicians from Singapore-based institutions KK Women’s and Children’s Hospital and the National University of Singapore used the system on 100 women who needed spinal anesthesia at KKH from May 2016 to May 2017.
Ninety-two percent of anesthesiologists were able to insert the needle on the first attempt during spinal anesthesia. The high accuracy rate reduces complications.
The researchers and physicians said the system can be helpful in training anesthesiologists to “better identify correct spinal landmarks.”
The research team sold the system to a company affiliated with the National University of Singapore, which will begin selling the system.