Global AI Impact Commons
AI in Radiology
Russia
Healthcare
Implementing Organisation
Celsus AI (Medical Screening Systems llc)
Russia, Moscow, Moscow
Implementing Point of Contact
Nikita Nikolaev
CMO
Contributor of the Impact Story
Carnegie India
Year of implementation
2017
Problem statement
The increased burden on radiological departments leads to high costs for treatment and rehabilitation, reduces the detection of oncological and other pathologies during screening.
Impact story details
Celsus is a developer of artificial intelligence-based solutions for X-ray and CT scan analysis. It has been developed and implemented by Medical Screening Systems LLC since 2017. Celsus has ready-made solutions for the analysis of mammograms, X-rays and fluorograms of the chest organs, CT scans of the brain and chest organs. According to the results of implementations and pilot projects, the use of Celsus not only reduces the time spent on reading and interpreting research by up to 50%, but also reduces the risks of errors related to the human factor. Celsus solutions are used in 44 regions of the Russian Federation. The company's services are successfully used in hundreds of medical institutions in Moscow and the regions of the Russian Federation and over the past 3 years have successfully helped doctors analyze and interpret more than 11 million studies. Celsus became the first Russian solution based on artificial intelligence in healthcare to receive a medical device registration certificate for the highest risk class. There is a patent for the invention and registered trademarks, the certification of the European Union (CE Mark). The solution is included in the register of domestic software. The company is a resident of the Skolkovo Foundation. Celsus is a winner and finalist in a number of prestigious competitions, such as: Amazon AI award, Digital Peaks, Roscongress Business Priority, AI award, CIPR, Data fusion, Data award, Brics award.
AI Technology Used
Key Outcomes
Efficiency
Productivity, Resilience
Risk Reduction
Narrative Outcome
Radiological departments face mounting pressure from high caseloads and limited availability of specialists. Delays in interpreting scans can result in missed diagnoses and higher downstream costs. Celsus AI's radiology tools reduce interpretation time by up to 50 percent and lower human-related errors. Mammography reporting time has been cut roughly in half. For health systems where specialists are stretched thin, these efficiency gains translate directly into more patients screened and earlier detection of conditions.
Impact Metrics
Time spent on reading and interpreting medical reports
Baseline Value
7.45 minutes for mammo reporting, 0.8 minutes for precision reporting
Post-Implementation
Reduces reading and interpretation time by up to 50% and lowered human-factor errors. Reduced mammography reporting to 3.55 minutes, with 0.84 minutes for precision reporting. Overall workload reduction ranges from 50-60%. Minutes
Implementation Context
The company's solutions are industrially used in the Russian Federation and occupy more than 40 percent of the existing market. Globally, solutions are being used industrially in the Islamic Republic of Pakistan and the Republic of Belarus. The solution supports full localization and can be deployed on the customer's server hardware. The company implements pilot projects in 6 countries.
Patients in need of TB and Cancer Screening
Key Partnerships
Moscow Department of Information Technology, Skolkovo Foundation, and Ministry of Health of the Russian Federation
Replicability & Adaptation
Technical resources would be required
Supporting Materials
* The data presented is self-reported by the respective organisations. Readers should consult the original sources for further details.