Artificial intelligence in medical imaging
> 95% accuracy *
*These data are based on the control group screening tests that included 400 patients. Groups for screening tests conducting were randomly formed from verified images not used in neural network training. Each image was re-read by radiologists.
The system detects malignant and benign neoplasms, calcium deposits, lymph nodes, fibrocystic breast changes, dense breast tissue (ACR). It interprets BI-RADS analysis results.
93% accuracy *
*These data are based on the control group screening tests that included 347 patients. Groups for screening tests conducting were randomly formed from verified images not used in neural network training. Each image was re-read by radiologists.
The system analyses photofluorographies to detect the presence or absence of pathologies.
Improving the quality and standardization of the work of departments of radiation diagnostics
Increase in the rate of detection of cancer in the early stages
Reducing the cost of diagnostic measures
Retrospective analysis of radiological data
Minimizing the risks associated with the “human factor” in the work of doctors
Compensation for the shortage of qualified personnel
To operate the system there is no need in purchasing any additional equipment or conducting long-term staff training.
Pilot launches and trial operation
Kaluga, Tver, Bryansk, Kursk, Nizhny Novgorod, Murmansk, Ivanovo, Tambov, Kaliningrad regions, Republic of Dagestan, Kabardino-Balkarian Republic, Moscow, St. Petersburg
Research Cooperation Memorandum
Nenets autonomous district
Republic of Kalmykia
Republic of Adygea
Republic of Bashkortostan
Republic of Komi
Republic of Dagestan