Network Security Internet Technology Development Database Servers Mobile Phone Android Software Apple Software Computer Software News IT Information

In addition to Weibo, there is also WeChat

Please pay attention

WeChat public account

Shulou

The combination of AI and CT can detect early esophageal cancer for the first time, and the new technology of Aridamo Hospital is expected to be on the ground in routine physical examination.

2025-04-04 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > IT Information >

Share

Shulou(Shulou.com)11/24 Report--

CTOnews.com, January 6, according to the official news of Ali Dharma Hospital, the Dharma Hospital joined hands with a number of domestic hospitals to combine AI with ordinary plain scan CT for the first time, which can effectively identify early esophageal cancer, with a sensitivity of 93% and a specificity of 98%.

The Aridamo Hospital said that esophageal cancer is one of the high incidence cancers in China. Data show that there were more than 600000 new cases of esophageal cancer and more than 370,000 deaths in the world in 2020, half of which came from China.

The high mortality of esophageal cancer is mainly due to the lack of low threshold and high reliable early screening methods in clinical medicine, and the difficulty of treatment is also greatly increased after the spread of lesions in the middle and late stages.

The Dama Hospital Medical AI team, together with the first affiliated Hospital of Zhejiang University Medical College, the Cancer Prevention and treatment Center of Sun Yat-sen University, Shengjing Hospital affiliated to China Medical University, Sichuan Cancer Hospital and other institutions, cut into this problem from the beginning of 2022.

The basic idea is to let patients only do plain CT with low threshold, and then use AI technology to identify whether there is an esophageal tumor, whether it is benign or malignant. Plain scan CT is the most common CT scan, which does not require additional injection of iodine contrast medium, and has been widely used in all kinds of physical examination.

In view of the fact that the size of early esophageal cancer is small and indistinguishable from normal tissue, the research team focuses on improving the segmentation algorithm model, introducing global attention mechanism, combined with location embedding, so that the segmentation model pays attention not only to the part, but also to the whole.

The blue line in ▲ map is AI recognition accuracy, that is to say, AI needs to learn various global features such as esophageal shape and texture, judge whether there are abnormalities such as asymmetric esophageal wall thickening or extrusion, and also analyze local image details. This humanoid algorithm design greatly improves the ability of AI to identify early esophageal cancer.

CTOnews.com learned that the relevant research papers have been included in MICCAI 2022, the international top meeting of medical imaging, and are currently undergoing multicenter large-scale clinical verification, and are expected to be widely landed in the physical examination project in the future.

IEEE Fellow Lu Le, head of the Dharma Hospital Medical AI team, said that this technology has opened the API call interface on the public cloud for doctors to experience and use, and is expected to be used in routine physical examination programs in the future, lowering the threshold for screening for esophageal cancer and achieving early diagnosis and early treatment.

API call API document: click here to view

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.

Views: 0

*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.

Share To

IT Information

Wechat

© 2024 shulou.com SLNews company. All rights reserved.

12
Report