Chinese researchers have found a new way to diagnose ovarian cancer without invasiveity

reporter from
Suzhou Medical Institute was informed that the researchers and Fudan University affiliated Jinshan Hospital and other 8 hospitals, the latest development of a new non-invasive diagnosis of ovarian cancer. Using this method, medical personnel can accurately judge the classification of ovarian cancer by using only the patient's MRI image.
this time, the research team, in cooperation with eight hospitals in East, South and North China, took the lead in applying artificial intelligence technology to imaging diagnosis of ovarian cancer patients. The researchers collected nuclear magnetic imaging data from 501 patients with ovarian cancer in two major categories, screened features and modeled them through machine learning methods, and finally formed a new method of non-invasive diagnosis of intelligent ovarian cancer. Comparing the machine diagnosis conclusions of the new method with 6 imaging doctors who worked for 2 to 13 years, the results showed that the average diagnostic accuracy of doctors was 79.5% and that the average diagnostic accuracy of machines was 91.7%. The diagnostic accuracy of the new method is significantly higher than that of labor.
" combination of artificial intelligence and imaging diagnosis has changed the thinking of manual reading diagnosis in the past. The new method not only helps doctors clinically, improves the accuracy of ovarian cancer diagnosis, but also suggests which factors are most valuable for forming diagnosis and helps doctors improve the efficiency of imaging examination. Gao Xin, a researcher at
Institute of Medicine and Technology in Shanghai, said.
results were published in February in the journal Magnetic Resonance Imaging, an authoritative journal in the field of radiology. (Source: Xinhua News Agency, )
relevant paper information: