Scientists in St. Petersburg have come up with a new artificial intelligence solution to support physicians. The system is expected to help doctors analyze mucous membranes in human organs. With the system physicians will be able to perform mass screening studies and dramatically improve examination of patients who live in far-flung regions where there’s little or no high-tech medical equipment. The AI comes from researchers at Peter the Great St. Petersburg Polytechnic University (SPbPU) who’ve teamed up with colleagues from the local Almazov National Medical Research.

The St. Pete artificial intelligence is aimed at aiding physicians in identifying an array of diagnostic states of an organ.

One of the key advantages of the system is reported to be the possibility to fully automate diagnostics during mass screening examinations. With the system, no subjective assessment of a medical image will be possible, thus excluding the human error factor, the developers said, adding that the AI would also enable high-quality medical consultations for people in remote and hard-to-reach areas.

The scientists do full analysis of an image and then mathematically describe various parameters.

“In the system, we apply deep neural networks that keep learning and evolve to isolate and classify pathologies on endoscopic images. The algorithms identify suspicious areas and show the probability of developing various pathologies,” said Vitaly Pavlov of SPbPU.