[ad_1]
With over 2,000 different skin diseases, it is impossible for non-specialists to know them all. That's the problem Shi Jiang has been facing for two decades as a general practitioner in a community health clinic in Shanghai, where most of his patients are elderly
"Sometimes they showed me their aging spots or moles , worried if they were abnormal.But it's been a long time since I did not study dermatology at university, says Shi, 45.
When he heard last month on Quality Skin, an experimental application for dermatological diagnosis powered by AI, Shi jumped
In China, health care has become one of the most recent applications of AI, with the security and the new retail trade, where technologies such as natural language processing and computer vision are being applied, greatly improving efficiency in a country where there is an acute shortage of doctors and qualified nurses. The e Chinese companies, from startups to technology giants, are seizing the opportunity to apply AI solutions to all applications, from automatic reading of CT scans to processing and editing. analysis of medical requests.
head of research at China-Japan Friendship Hospital, which was chosen by the authorities to experiment with medical consultations on the Internet and created the Quality Skin application that can produce weighted results in one minute by receiving high-resolution images of "There are more than 2,000 types of skin diseases, while new graduates can only recognize 100, and even the most experienced doctors will only find 800 in their entire careers ", he said. "Most doctors avoid giving their opinion on an illness that they do not know, and 60% of cases are at risk of misdiagnosis."
While only 7.7% of Chinese hospitals are rated "triple A", they processed half of all outpatient visits in 2016, according to industry estimates cited in the recent Ping An Prospectus Doctor. Most doctors only receive early training and do not have the opportunity to follow new medical developments.
In China, the dermatology departments of hospitals and clinics account for about 3% of the 240 million patients. "China is already far ahead of the US in AI diagnostics," said Ni Hao, general manager of healthcare at Yitu Technology, a Shanghai-based AI start-up specializing in facial recognition systems. There are two reasons for this, according to Ni. First, the shortage of doctors in the United States is not as serious as in China, and second, US start-ups can not afford to have a large medical team because the US doctors earn between $ 250,000 and $ 300,000 US.
Yitu supports a team of about 400 doctors, most of whom work part-time for about 10 hours a week to help with data labeling, while a fifth of his employees at full time have a medical history.
The company has partnered with Chengdu's West China Hospital on an AI-driven lung cancer diagnostic system that can tap approximately 280,000 actual cases of lung cancer – the largest single base of data of the kind in the world – and
Yitu is one of dozens of Chinese start-ups that have emerged in recent years, and many are now focusing on health care. In November, a smart robot developed by iFlyTek, a listed company in Shenzhen, passed the country's national written qualification examination for doctors.
The AI health opportunity has not been overlooked by the country's biggest tech giants. The ubiquitous WeChat messaging platform was selected by Beijing last November to take the lead in AI applications in healthcare. He has since teamed up with more than 100 triple A hospitals across the country.
Last week, the Shenzhen-based company launched its first open platform for medical diagnosis AI. With the help of computer vision and AI analysis, the Tencent Shadow Chaser covers 700 diseases that make up 90% of the most common outpatient cases.
Similar to the training of doctors, the platform went through three stages: "First, natural language processing and in-depth learning of medical textbooks, health records and directories. diagnosis to produce a stock of medical knowledge, followed by the development of diagnostic algorithms and optimization algorithms with the help of human medical specialists. 19659002] Shadow Chaser, which was included in about 10 million databases on health, is a step towards building the next generation of intelligent health services, said Vice President Tencent Chen Guangyu in a statement.The Baidu engine migrated to open source AI technologies that can help pathologists detect breast cancer, while the Alibaba group has announced reconciliations with hospitals for intelligent diagnostic platforms. n October, Alibaba Health launched its first AI medical laboratory in cooperation with affiliated hospitals of Zhejiang University and Xinhua Hospital. He also announced a blockchain-enabled public platform to serve as a secure data sharing network for hospitals.
The China-Japan Friendship Hospital, which launched in March an AI-assisted diagnostic platform for skin cancer, has received more than 10,000 requests from doctors wishing to be diagnosed with the disease. adopt automated consultations. "There are more than 20,000 patients diagnosed in China every year, but many of them do not find out until too late," he says.
The algorithm is now mature, and it is only a matter of time before we can transfer what the machine has learned in the diagnosis of other diseases of the skin, "said Cui, adding that the team is working to add a new skin disease every three months to the platform
.No diagnostic services authorized by the country's AI has received approval from the Food and Drug Administration of China (CFDA) – a prerequisite for commercial adoption.The assisted diagnostic systems of West China Hospital and China-Japan Friendship Hospital are experimental and currently free, so they do not require the approval of the authorities.
"We want to show the industry what the diagnosis and treatment of AI allow The system will like, "said Ni de Yitu, adding that if his lung cancer program is known sceptible to be approved next year, the process has been "difficult to manage" due to regulatory requirements.
Even with the delay in regulation, patients benefit. Shi, the Shanghai doctor, was able to identify a potentially malignant condition in a 68-year-old patient after the Quality Skin application indicated 70% chance of the most common form of cancer. the skin. Since then, he uses it every day.
"Once you start working as a generalist, it's hard to keep track of the different disciplines," said Shi. – Message from South China
[ad_2]
Source link
