Russian specialists from Skoltech have trained a neural network on the optical sensitivity of the lungs, which is extracted from the data obtained and describes them, in order to perform all the tasks that are currently assigned to doctors. The precipitation study has already been noted in Nature Scientific Reports.
As assemblers, by applying the new solution, the image analysis time will be reduced in a few minutes and, as a result, will be no more than 30 seconds, after which the need arises or the diagnosis assumed by the neural network is confirmed or any corrections are identified.
The training of the neural network from Skoltech is still based on the “image-text” pair and, therefore, it is able to make only those diagnoses that it “knows” and which can be determined from the images. However, due to the models of machine medicine and computational linguistics, her training continues.
In addition, the “knowledge” of the neural network in this area is supplemented by a radiological dictionary, which makes it possible to make more accurate diagnoses in terms of radiological terms, and a consolidated sample of the sample intended for study is being formed.
In the near future, active interaction of neural networks based on the corrections carried by the doctor during the diagnosis is expected, as well as the study of the possibility of using a similar neural network for processing MRI and CT data. research.
Source: Tech Cult
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
