Taiwan's smart medical care has taken a big step forward. National Taiwan University Hospital, Cathay General Hospital and medical imaging AI startup Yunxiang Technology announced today (27th) that the three-party cooperation "colonoscopy real-time AI polyps aetherAI Endo" has obtained the TFDA medical material license from the Ministry of Health and Welfare. AI-assisted screening can improve the detection rate of colonoscopy polyps. The accuracy of this AI model is as high as 95%, and China Tai Hospital plans to be the first to introduce it in the health examination center.
20220927 National Taiwan University Hospital, Cathay General Hospital and Yunxiang Technology cooperated with each other to produce real-time AI polyp detection smart medical materials for colonoscopy certified with_0.
National Taiwan University Hospital, Cathay General Hospital and Yunxiang Technology collaborated to certify real-time AI polyp detection smart medical materials for colonoscopy.
Picture/Yunxiang Technology
Cross-institutional collaboration to create “hardware and software integration” AI tools
According to statistics from the National Health Insurance Administration in 2022, there are nearly 17,000 new cases of colorectal cancer in Taiwan every year. Wu Mingxian, president of National Taiwan University Hospital, pointed out: "Colorectal cancer is a new national disease, with the highest incidence rate." Colonoscopy has been proven to be effective. Reduce the mortality and incidence of colorectal cancer. However, doctors have different abilities to detect adenomas during colonoscopy, which can easily cause the problem of missed lesions.
Qiu Hanmo, director of the Department of Health Management and Comprehensive Diagnosis and Treatment of National Taiwan University Hospital, pointed out that the indicator of colonoscopy is the "adenoma detection rate." According to research, every 1% increase in the adenoma detection rate can reduce the risk of colorectal cancer after microscopic examination by 3%. Colonoscopy through AI human-machine collaboration can increase the adenoma detection rate by approximately 14 to 30%.
Stay on top of the latest AI, semiconductor, and digital trends! Subscribe to the "Digital Times" daily newspaper and social event information
Please enter Email
Colonoscopy real-time AI polyp detectionaetherAI Endo
Picture/NTU briefing
Yunxiang teamed up with National Taiwan University Hospital and Cathay General Hospital, and it took four years from development to evidence collection. It created nearly 200,000 imaging data from 3,000 patients at National Taiwan University Hospital and Cathay General Hospital, and after annotation by gastroenterologists, it established AI detection model with an accuracy of up to 95%.
Ye Zhaoyuan, CEO of Yunxiang Technology, said that unlike the ivory coast telegram past development by a single institution, this time it is a collaboration between two medical centers and technology companies, which has important indicators. Since personal information laws strictly protect patient privacy, hospitals need more diverse data to build AI models to be more comprehensive.
This set of smart medical equipment combines software and hardware, and the installation process is quite painless. The colon endoscopy instrument host in the hospital will be plugged into an AI Box computing host provided by Yunxiang and connected to a screen. When the AI detects polyps, an alarm sound will be emitted to assist the doctor in interpretation. The computer and software are all made in Taiwan, and it only takes 0.1 seconds to make a diagnosis and alert the doctor," said Ye Zhaoyuan.
Qiu Hanmo pointed out that 400,000 colonoscopy examinations are performed every year, which is quite frequent. The quality of colonoscopy examinations is improved through AI assistance. "In the future, people will not have to gamble when entering a hospital and have to ask which one has better quality."
- Board index
- All times are UTC
- Delete cookies
- Contact us