The Big Data research team built a software test to automatically diagnose the disease based on the input data as X-ray images, with up to 90% accuracy.
The software is built based on open data from actual X-ray images of both ill and non-ill patients, then for machine learning and identification. The results show that based on the evaluation criteria of the health sector, both sensitivity and specificity of the algorithm reach over 90%, even over 95% for some aspects. If you rely on processing speed criteria, to read a standard X-ray image doctor will need 5-10 minutes but the reader takes only 2-5 seconds depending on hardware configuration.
Professor Vu Ha Van, Director of the Big Data Research Institute (Big Data Institute - Vingroup), said: "To be sure of our ability to carry out international standard research, the Analysis team Medical image of the Institute has sent the software to participate in 2 prestigious competitions, held on an open platform.In the diagnostic diagnosis of pneumothorax segmentation by the American Medical Imaging Association Organized by SIIM, the Institute's research team is in the top 5 positions among over 1,000 teams competing from around the world, including many strong teams from prestigious organizations of Russian Federation, United States, China, Israel ...
