AI model diagnostic accuracy comparable to experienced pediatricians
February 13, 2019 Source: Science and Technology Daily Author: Zhang Meng Ran
Window._bd_share_config={ "common":{ "bdSnsKey":{ },"bdText":"","bdMini":"2","bdMiniList":false,"bdPic":"","bdStyle":" 0","bdSize":"16"},"share":{ }};with(document)0[(getElementsByTagName('head')[0]||body).appendChild(createElement('script')) .src='http://bdimg.share.baidu.com/static/api/js/share.js?v=89860593.js?cdnversion='+~(-new Date()/36e5)];According to a recent study published online by the British journal Nature Medicine on the 12th, American scientists reported a highly accurate artificial intelligence (AI) model that is equivalent to an experienced pediatric diagnosis of common childhood diseases. Doctors. The findings confirm, in a proof-of-concept manner, that artificial intelligence-based systems can assist pediatricians in processing massive amounts of data, enhance diagnostic assessments, and provide clinical decision support in cases where the diagnosis is unclear.
In recent years, with the increasing complexity of medical information, the range of diseases, diagnostic tests and treatment methods have grown exponentially, and clinical decision-making has become increasingly complex.
This time, Zhang Kang, a researcher at the University of California, San Diego, and his colleagues developed an artificial intelligence-based model that uses an automated natural language processing system to identify clinically relevant information from electronic medical records through deep learning techniques. The model can search for electronic medical records and discover associations that were not discovered by previous statistical methods.
The team trained and validated the framework by analyzing 101.6 million data points in 1,362,559 pediatric outpatient clinics in a large referral center in Guangzhou, China.
Compared with the initial diagnosis results of the examining physician, the model has high diagnostic accuracy and is particularly good for the diagnosis of two important diseases: one is common diseases such as influenza and hand, foot and mouth disease, and the other is acute asthma attack and meninges. a serious or life-threatening condition such as inflammation.
Researchers believe that such artificial intelligence frameworks can be differentiated by patient classification, distinguishing patients who may be just a common cold from critically ill patients who need urgent intervention, thereby improving patient care efficiency.
Last year, Google’s artificial intelligence subsidiary DeepMind has successfully trained an artificial intelligence algorithm to detect eye diseases through massive data, and claimed that it is faster and more efficient than human experts, and will help humans eliminate “can Avoid blindness."
Polycarbonate Greenhouse,Po Film Covering Greenhouse,Venlo Type Glass Greenhouse,Multi Span Plastic Film Greenhouse
Qingzhou JuLong Agriculture Equipment Co.,Ltd , https://www.julonggrowtech.com