Network for metastases
Russian scientists have developed a neural network to search for cancer metastases
The Russian company Unim has trained a neural network developed by it to detect cancer metastases in the human body, the company said in a statement.
Metastasis is a distant secondary focus of the pathological process. It is necessary to identify metastases in order to correctly determine the stage of cancer, otherwise the patient will receive insufficient treatment.
As noted in the company, the neural network determines in 10 seconds whether there are formations similar to metastases in the studied material and where they are located. The doctor spends at least half an hour on this.
To create a neural network, more than 60 terabytes of diagnostic data were analyzed, dozens of pathologists participated in its training. While the system is working in test mode, it has been used to check the data of 600 patients who already have a diagnosis.
"Neural networks take histology to a whole new level. In the next couple of years we will see a new level of evidence and accuracy of oncological diagnosis. The doctor will have a reliable assistant who will significantly reduce his labor costs and significantly increase work efficiency," the words of Unim CEO Alexey Remez are quoted in the message.
Unim has been operating since 2014. She is developing new tools for the diagnosis of cancer. The company's investors are FRII, Medme, as well as Deputy CEO, Investment Director of the Russian Venture Company (RVC) Alexey Basov. The company is a resident of the Skolkovo Foundation.
According to the International Agency for Cancer Research, Russia in 2018 ranked fifth in the world in terms of the number of deaths of cancer patients. According to the Ministry of Health, the number of cancer patients in Russia in 2017 reached 3.5 million people, about 600 thousand Russians receive an oncological diagnosis a year, the report says.
Portal "Eternal youth" http://vechnayamolodost.ru