Human-machine war"! Artificial intelligence installs "strongest brain" for genetic medicine

CNR Zhengzhou, December 14 (Reporter Hu Xiaohui) On the afternoon of December 13, a "human-machine battle" organized by the Genetics and Prenatal Diagnosis Center of the First Affiliated Hospital of Zhengzhou University was held in Zhengzhou: five professional researchers and A unique PK competition was carried out between the intelligent analysis platforms of human chromosomes. In the end, the artificial intelligence won. It took an average of 377 seconds to look at a karyotype manually. The machine completed an average of 1 karyotype in 0.78 seconds. The artificial intelligence was 483 times the manual analysis.

Professor Kong Xiangdong, director of the Prenatal Diagnosis Center of the First Affiliated Hospital of Zhengzhou University, said that from the results of the competition, it can be seen that "artificial intelligence" has fast speed and high artificial precision. The combination of the two and applied in clinical practice will be the best in the future. s Choice.

The karyotype analysis work is mainly used to solve the cytogenetic karyotype analysis problems encountered in medicine. In the past, chromosome karyotyping work was handled by observing chromosome slides under a microscope. For ordinary people, each chromosome is just an identical "bug" under the microscope. Doctors need to observe and record under the microscope for a long time, which is a great test for the doctor's cervical spine and eyes. Long-term work will cause serious damage to the doctor's health. Moreover, each type of chromosome has specific characteristics. To distinguish them, it takes at least 3 years to develop a skilled testing researcher. The talent training cycle is long, the cost is high, and the uncertainty is high.

"In our genetic center, we have continuously introduced new technical means to improve the working method, gradually improving the working efficiency and reducing the working intensity. The working mode has gradually changed from facing the microscope to facing the computer screen," said Kong Xiangdong. The way of working mainly relies on the doctor's eye identification and manual mouse clicks to count, but the work is mechanical and cumbersome, and has very high requirements on the doctor's eyesight, physical strength, and energy.

The intelligent analysis platform participating in this competition was developed and researched by the Prenatal Diagnosis Center of the First Affiliated Hospital of Zhengzhou University and Hunan Zixing Artificial Intelligence Technology Co., Ltd. After more than a year of technical research, it officially entered the clinical stage. With intelligent assistance, doctors in the Genetic and Prenatal Diagnosis Center can quickly and accurately complete the intelligent detection of chromosome karyotype analysis. By continuously introducing new technologies, the efficiency is greatly improved and the time for patients to wait for results is reduced. Currently, the First Affiliated Hospital of Zhengda University The karyotype analysis of peripheral blood in the Genetic and Prenatal Diagnosis Center has achieved the fastest speed in the country. At the same time, it also reduces the workload of doctors in tedious work, allowing doctors to concentrate their energy and knowledge to solve other problems in genetics, so that precious medical resources can be used more effectively.

"I believe that the era of artificial intelligence + medical care has come." Professor Kong Xiangdong said, "medical care + artificial intelligence" is constantly promoting the modern medical industry to upgrade to future medical care. In the future, artificial intelligence, big data, etc. and economic and social development will go deeper The integration of levels, "Artificial intelligence provides new processing methods and medical perspectives. The implementation of artificial intelligence in the medical field requires a large amount of medical data as support, and we have saved a large amount of medical data for decades. How to effectively use this batch of medical data? Data, converted into machine experience, this will be the next direction we explore with artificial intelligence companies."