Ping An Technology repeats success
ISBI, the world's top medical imaging conference, recently released its 2019 international competition results.
Ping An Technology, a subordinate of Ping An Group, won the championships in three competitions for Automatic Lung Cancer Segmentation and Classification (ACDC), endoscopic image quality control (EAD) and pathological myopia challenge (PALM), as well as six sub-tasks.
This is another achievement made by Ping An after winning the first, second and third place of EX (hard exudate), HE (hemorrhage) and MA (microaneurysm) segmentation tasks respectively in the fundus analysis competition at the ISBI conference last year.
This means that Ping An has once again achieved superior results in AI-assisted medical technology and laid a solid foundation for the development of the Ping An medical health eco-sphere.
The above three competitions have significant clinical application values in the medical field.
High-quality imaging will increase diagnostic accuracy and significantly improve the efficiency of doctors in reading imaging. It also has higher application significance for the development of computer-aided ergonomics diagnosis tools and early screening tools.
ISBI is an international academic conference dedicated to biomedical imaging initiated by the IEEE Signal Processing Association (SPS) and the IEEE Biomedical Engineering Association (EMBS). Every year, ISBI holds an international competition for medical imaging knowledge, attracting world-renowned universities and research institutions.
With the accumulation of technologies in the medical knowledge map of millions of medical concepts, professional medical imaging knowledge, and the AI algorithm platform of Ping An Medical Technology, Ping An surpassed opponents to get first prize in the international competition.
 |
|
Ping An Technology is attracting wide attention with its technological innovations. [Photo provided to chinadaily.com.cn] |
In the ACDC competition, Ping An won the championship with a total score of 0.8373, defeating 391 participants from South Korea's Lunit, Beijing University of Aeronautics and Astronautics, and the National Cancer Institute.
With the AI automatic detection of Endoscopic artefacts as the match content, Ping An Medical Technology won first place with a total score of 0.3610 in the EAD Quality Control Competition.
The team also reaped laurels in such sub-tasks as endoscopic multi-category artefact detection, endoscopic artefact segmentation and endoscopic detection generalization, triumphing over competitors from the Technical University of Munich, the University of California at Davis, Osaka University, King's College London, Hikvision and more.
When it comes to the detection of pathological myopia, which requires AI algorithms to detect pathological myopia in retinal fundus images, Ping An Medical Technology won out over other participating teams with three leading core indicators.
The company also won the two sub-tasks of optic disc and lesion segmentation, defeating 345 participants from Tsinghua University, Chinese University of Hong Kong, Shenzhen University, the National University of Singapore, the University of Leuven in Belgium and more.
Xiao Jing, the chief scientist of Ping An Group, remarked that lung cancer is the most common malignant cancer in the world, the pathological segmentation of which is highly clinically relevant.
The automatic segmentation of cancerous lung nodules from pathological sections through AI-assisted system helps to reduce the workload of pathological doctors and subjective bias in diagnosis, which are of great significance for the auxiliary diagnosis and analysis of lung cancer.
Xie Guotong, the chief medical scientist of Ping An Group, mentioned that endoscopy is widely used in the early detection of gastric and colon tissues, therefore endoscopic artefact examination is of great significance for medical imaging quality control and clinical diagnosis.
Pathological myopia belongs to the retinal degenerative group of diseases, the incidence of which among people with high myopia is increasing year by year, so the early diagnosis and regular follow-up to treat this condition is very important.
In recent years, relying on its advanced AI technologies, Ping An Group has continued to promote the strategies of “finance + ecology” and “finance + technology” and build a strong medical knowledge base and health-care ecosystem.
The group has integrated an end-to-end solution matrix in fields such as intelligent medical quality control, intelligent disease prediction, intelligent triage, clinical intelligent image screening, and AI-assisted diagnosis and treatment.
It starts with medical treatments before, during, and after treatment to help solve medical pain points faced by patients, doctors and medical institutions.
The medical AI model in auxiliary diagnosis and treatment launched by Ping An Group has covered over one thousand common diseases, which has benefited around 800 million potential patients.
Ping An Smart City - Wise Medical has developed close cooperation with medical administrative authorities in nearly 20 provinces and cities, signing contracts with over 300 medical institutions.
