ICLR (International Conference on Learning Presentations) is one of the three most prestigious international conferences in the field of machine learning. In order to promote the development of machine learning technology in the medical and health field, ICLR 202 1 launched this competition. The theme of this competition is the development of automatic medical diagnosis dialogue system, which is the first international challenge for the generation and diagnosis of medical dialogue. There are two main tracks: medical dialogue generation track and disease automatic diagnosis dialogue system track. In the end, Iflytek Medical successfully won the championship of automatic medical diagnosis dialogue system.
Iflytek Medical can successfully win the championship this time, which is inseparable from the word "quasi". Its automatic medical diagnosis dialogue system can simulate the dialogue between doctors and patients with words. Firstly, the system predicts the disease according to the symptoms reported by patients themselves. Subsequently, the system will predict the next symptom to be asked according to the pre-judgment result, and so on until the stop condition is met; Finally, all the obtained symptoms will be sent to the automatic diagnosis system for disease prediction.
In medical diagnosis, the accuracy of "question" and "diagnosis" directly determines the effect of consultation. During the contest, Iflytek passed the test of 12 disease, 1 18 symptoms and a total of 2,374 dialogue data, and finally won the championship among 73 teams in the world, achieving the "first in the world" diagnostic accuracy.
As the world's leading artificial intelligence enterprise, Iflytek has made remarkable achievements in the field of smart medical care in recent years. Through the construction of artificial intelligence aided diagnosis and treatment system, Iflytek AI technology deeply cuts into the doctor's clinical diagnosis process, and gives auxiliary diagnosis suggestions during the doctor's diagnosis process, thus improving the diagnosis and treatment ability and service level of doctors, especially grassroots doctors.
At present, Iflytek A.I. Science and Technology has made remarkable achievements in the fields of people's livelihood, such as standardizing the behavior of primary diagnosis and treatment and improving the level of primary medical services. According to Iflytek's financial report in 2020, based on the innovation of "depth model+knowledge reasoning", Iflytek Intelligent Medical Assistant provides medical records quality control, auxiliary diagnosis, rational drug use, medical knowledge retrieval and other functions for grassroots medical staff, which has been well applied. In 2020, Iflytek Medical Assistant will realize the comprehensive coverage of primary medical institutions in Anhui Province, and gradually start to be applied in Tibet, Inner Mongolia, Qinghai, Beijing, Xinjiang, Heilongjiang, Zhejiang and other places. It has been launched in more than 30,000 primary medical institutions in more than 70 districts and counties nationwide, serving tens of thousands of primary doctors.
At present, there are still structural problems in China, such as the relative lack of high-quality medical resources and the insufficient capacity of primary medical services. Iflytek is constantly exploring the use of science and technology to empower primary health care, which provides an effective way to improve the work efficiency and service ability of medical staff. It is believed that with the continuous expansion of the application scale of smart medical care, the high-quality medical resources in China will be distributed more fairly and efficiently.