Deep Application Of Artificial Intelligence Landing Hospital Pathfinding

Artificial intelligence has been applied in a variety of scenarios in hospitals, which has effectively improved patients' medical experience in terms of convenient patient registration, shortened waiting time for medical treatment, and assisted diagnosis. The reporter learned that artificial intelligence has been widely used in hospitals before, during and after diagnosis. But at the same time, industry insiders also said that there are still many problems to be solved in the further application of artificial intelligence in the medical field.

Improve the utilization efficiency of hospital resources

"Health is the most value and significance of artificial intelligence application areas, to promote medical intelligent upgrade is the key to improve medical soft power, has been in Shanghai will be the application and development of artificial intelligence in the medical field as the key link of industry layout, actively explore AI + ', 'medical treatment, to build research and development - industry - application" fast track ", To explore and try to drive the digital transformation of medical care and the precise development of medical care." Sun Yue, deputy director of the artificial intelligence Development department of the Shanghai Municipal Commission of Economy and Information Technology, said at a forum on the "high-quality development of human and Industrial Intelligence Enabled hospitals" held recently.

At present, artificial intelligence has been well applied in the aspects of convenient medical treatment, improving patient experience, assisting diagnosis and treatment, improving the efficiency of hospital resource use, and improving the fine management level of hospitals.

"Registration is easier, the doctor's professional alignment, a short appointment time, less line up, less 'wrong way' to see a doctor, is the common appeal of patients." Wang Yu, president of Shanghai First Maternal and Infant Health Care Hospital, introduced that the hospital has launched many new functions through AI, which help to improve the efficiency of hospital resources.

"Take the 'doctor seeking helper' function launched by us as an example. According to the symptoms complained by patients, it can match the corresponding doctor resources for patients through the combination of AI robot and human customer service, so that patients can register the 'right' number. In view of the unfamiliar situation of the first medical treatment process for patients, we have opened the intelligent guidance of the medical treatment process, so that the "patients seeking services" in the past is now "service seeking people". In order to reduce the waiting time for patients to see a doctor, we have also introduced functions such as "online card establishment" and online examination sheet issuance. At present, the interview session of our on-site doctors has been reduced from 3 times to 1 time. "The whole time for building the obstetric card has been reduced from 4 hours to 1.5 hours." Wang Yu introduced.

Hu Weiguo, deputy director of Ruijin Hospital affiliated to Shanghai Jiao Tong University School of Medicine, said that Ruijin Hospital also helps optimize the whole patient treatment process based on artificial intelligence technology, and accurately calculates the treatment time of different specialists and different experts based on big data, which has achieved the accuracy of the source time of all department numbers within 30 minutes.

Enabling medical diagnosis and development

Digital technologies represented by artificial intelligence have also brought changes to medical diagnosis and treatment, medical device research and development. "Taking the auxiliary diagnosis of chest CT as an example, at present, we complete the auxiliary diagnosis of chest fractures and lung nodules by AI in 80% of the daily life. The diagnostic time of 1mm imaging was reduced from 7 minutes to 2 minutes. In terms of coronary CTA assisted diagnosis, Ruijin has achieved a significant increase in the number of patients diagnosed with coronary artery every day. "Ruijin has also carried out AI-assisted diagnosis of cervical cancer and gastrointestinal tumor pathology, and the efficiency of detecting malignant cells has increased by two to three times." Hu Weiguo said.

Wang Yanfeng, assistant director of Shanghai Artificial Intelligence Laboratory, introduced that medical AI has developed rapidly in recent years. According to incomplete statistics, 12 domestic medical AI enterprises have obtained the medical device Class III certificate, and a total of 79 international AI systems have been approved by FDA.

In recent years, Shanghai shen-kang hospital development center in Shanghai to build artificial intelligence highland strategic deployment, with support from the relevant departments to guide, organization of Shanghai municipal hospital in the city, make full use of artificial intelligence, big data, 5 g digital technology can be used, such as convenient medical service digital transformation of 1.0 to 2.0 for the convenience of application scene construction, Artificial intelligence technology has been fully enabled in all medical links before, during and after diagnosis and treatment, and the quality of whole-range medical services integrated with Internet diagnosis and treatment has been improved, effectively improving patients' medical experience. At the same time, Shanghai Shenkang Hospital Development Center actively promotes the construction of the first medical big data training facility in China, helps the standardized and orderly development of artificial intelligence, and accelerates the application of medical artificial products in municipal hospitals.

Medical AI still faces many challenges

Although artificial intelligence is accelerating its introduction to the medical industry, industry insiders admit that medical AI still faces many difficulties. Wang Yanfeng said, on the one hand, large-scale open source data is the key to the rapid breakthrough of the current artificial intelligence in text, voice, image analysis, but the strict protection of medical privacy, it is difficult to large-scale open source; On the other hand, the current paradigm of supervised learning in the medical field only focuses on covering certain types of head diseases, and a large number of diseases with a long tail distribution have not been paid attention to.

Wang Yanfeng believes that the future development direction of medical field of artificial intelligence, is "the data distribution, modal, tasks, such as differences in collaborative modeling challenges, to explore the medical AI from special to general technical implementation", and "formation system based on the training model of open sharing, open source alternative to model data source, through the" base "model to help the long tail disease modeling, Accelerating the development of medical AI ".

Zhao Haipeng, vice president of Shanghai Tongji Hospital, said that as the application scenarios of artificial intelligence medical treatment become more and more perfect, the higher the dependence on artificial intelligence, the reliability will be the premise of the long-term development of the industry, and its importance is highlighted.

Wang Chunming, director of the intelligent medical development department at Renji Hospital affiliated to Shanghai Jiao Tong University School of Medicine, said that the cultivation of interdisciplinary talents capable of "artificial intelligence and medicine" needs to be strengthened. Taking textbooks as an example, the introduction of artificial intelligence is only a few chapters in medical textbooks at present, and relevant knowledge needs to be further systematized. Contents such as data label cleaning and application cases of artificial intelligence in medicine need to be enriched and supplemented.

In addition, Wang Chunming also said that in order to further accelerate the application of artificial intelligence in the medical field, it is necessary to form an "exploratory atmosphere" of artificial intelligence in hospitals, so that doctors can actively explore the application of artificial intelligence when facing clinical difficulties, and discuss the promotion with artificial intelligence enterprises.