NUH AI Interprets Spinal Stenosis in ’47 Seconds’ and More AI Letters
National University Hospital unveils AI for the spine
National University Hospital has introduced an AI tool to analyze lumbar spinal stenosis, a common indication for surgery in elderly patients.
According to a press release, the tool, called Spine AI, automatically detects areas of spinal canal narrowing and categorizes them by severity. It was trained using more than 18,000 MRI images of the lumbar spine from 446 patients. In one study, it took as quickly as “47 seconds” to analyze each spine.
Lumbar spinal stenosis is a condition in which the spinal canal in the lower back narrows, pinching the nerves and blood vessels that supply blood to the lower extremities.
Spine AI, which is currently being evaluated in the radiology department at NUHS, was developed by NUH, together with the NUS School of Computing and the National University Spine Institute. Siemens Healthineers was also engaged to optimise the AI’s user interface.
NUH is expected to generate around 4,000 lumbar MRI scans per year. With Spine AI automating the analysis in an average of seven minutes per scan, it could save 466 hours of time per year, estimates Dr Andrew Makmur, NUHS Group CTO and a consultant to the NUH Department of Diagnostic Imaging.
Korea to use AI to monitor seniors living independently
South Korea’s Ministry of Health and Welfare and the Comprehensive Support Center for the Elderly Living Alone will use AI to reach out to approximately 42,000 seniors expected to spend the upcoming Chuseok holiday alone.
Chuseok is a major autumn harvest festival on the Korean Peninsula. In South Korea, it is celebrated for three days.
The AI Call service is part of the government’s personalized elderly care service and uses text-to-speech and speech-to-text technology to make phone calls and collect and transmit call information.
Meanwhile, SK Telecom and Lotte Welfare Foundation have been roped in to provide technical assistance for the holiday call service.
Thai regional hospital tests CXR AI
Phrapokklao Hospital, a major regional hospital in Chanthaburi, a province southwest of Thailand’s capital Bangkok, recently implemented an AI chest X-ray (CXR) system to improve lung cancer screening in the community.
“In most Thai government hospitals, (CXRs) are interpreted by non-radiologists. However, in community hospitals, there are often no radiologists available to read CXRs at all. By using specialized AI to read all cases, we can support clinicians in detecting incidental high-risk nodules that may lead to lung cancer,” said Dr. Passakorn Wanchaijiraboon, deputy director of Phrapokklao Hospital’s Cancer Centre of Excellence, who led a recent evaluation study of the CXR solution qXR by Indian company Qure.ai.
“The implementation of CXR AI is particularly useful in community hospitals, where it can significantly improve diagnostic capabilities in the absence of on-site radiologists,” adds Dr Wanchaijiraboon, who is also an oncologist.
Earlier this year, Qure.ai received approval from the US Food and Drug Administration for the lung nodule about the qXR range of CXR analysis solutions.