What’s New: Samsung Medison and Intel are collaborating on NerveTrack™, a real-time nerve tracking ultrasound feature that helps anesthesiologists identify nerves in a patient’s arm to help administer anesthesia quickly and accurately. Leveraging the Intel Distribution of OpenVINO toolkit for computer vision and annotation, Samsung Medison’s NerveTrack™ can potentially reduce scanning time by up to 30 percent.1
“NerveTrack can detect the median and ulnar nerve with reasonable accuracy over almost the entire range of the forearm. It can shorten inspection time, finding the nerves in real time — even if the doctor does not trace from the wrist to the proximal direction. In particular, it can detect the ulnar nerve even at a level where landmarks such as the ulnar artery are not adjacent. This kind of features could help safe needle procedure by allowing the nerves to be separated from the surrounding tissues and vessels.”
–Professor Jee Youn Moon, MD, PhD, Seoul National University Hospital
Why It’s Needed: Ultrasound-guided regional anesthesia (UGRA) is becoming standard practice for needle-based interventions, including vascular access and peripheral nerve block. However, even with UGRA, it can be difficult for anesthesiologists to correctly identify nerves, which can be as small as 2 millimeters in diameter, or to see the needle tip properly. NerveTrack can automatically identify nerves in real time for anesthesiologists — reducing the possibility of complications while improving workflows.
“To keep up with the changing world of healthcare, you need trusted partners and flexible technologies. That’s why we teamed up with Intel to create our NerveTrack solution. With our combined industry expertise and cutting-edge solutions, we’re using innovative technologies to help practitioners identify nerves faster and more accurately. The result is potentially less risk, better patient outcomes and more efficient workflows,” said Dr. Won-Chul Bang, vice president responsible for product strategy at Samsung Medison.
How It Works: NerveTrack was developed based on Intel’s OpenVINO toolkit. It uses inference to detect and identify the location of a nerve area in real time during an ultrasound scan, improving the treatment workflow for anesthesiologists. To train Samsung’s proprietary real-time algorithm that automatically detects nerves in ultrasound images, a significant amount of clinical ultrasound data was required. And with Intel’s OpenVINO CVAT (Computer Vision Annotation Tool), the total volume of training data increased up to 7 times, leading to improved accuracy of more than 20 percent.
“NerveTrack can help physicians automate mundane and time-consuming tasks and free them to spend more time with their patients,” said Alex Flores, medical imaging director, Health and Life Sciences at Intel. “We are working closely with Samsung Medison to help improve patient experiences and reduce physician workloads.”
More Context: Samsung Collaborates with Intel to Speed Nerve Detection and Improve Workflows (Solution Brief) | Intel AI Powers Samsung Medison’s Fetal Ultrasound Smart Workflow | Artificial Intelligence at Intel
1 Internal Samsung testing. System configuration: Intel Core i3-8100H CPU @ 3.0GHz, 8GB memory, OS: 64-bit Windows 10.