Artificial intelligence will help detect thoracic aortic aneurysm

December 7, 2021

Artificial intelligence-based algorithms can now recognize thoracic aortic aneurysm on radiographs. These services are available to medical diagnosticians as part of an experiment for the computer vision technologies implementation into Moscow healthcare system. The doctor will be able to pay attention to the smallest deviations and interpret images faster, so that the patient could go for consultation and treatment on earlier stages.

"Now the capital's doctors will be able to use artificial intelligence to diagnose thoracic aortic aneurysms. The importance of this disease early intelligence is based on its asymptomatic course and high risk of death in case of complications. The digital assistant allows it to draw the specialist's attention to the smallest deviations and interpret images faster. Thus, artificial intelligence services are now available to medical diagnosticians for disease intelligence in seven fields of research," said Anastasia Rakova, Deputy Moscow Mayor for Social Development.

She also noted that improving the quality of diagnostics and its accessibility for Muscovites is one of the most important tasks for the city. Last year, digital equipment was networked, and artificial intelligence-based services were implemented into radiologists practice. Anastasia Rakova recalled that initially the services helped analyze images only for signs of COVID-19, but gradually the capabilities of neural networks within the framework of the Moscow experiment expanded.

The services help doctors find signs of illnesses. Computed tomography of the chest organs can detect lung cancer, COVID-19, osteoporosis of the spine, thoracic aortic aneurysm, signs of coronary heart disease, radiography and fluorography — lung pathology, mammography — breast cancer.

The algorithms are currently undergoing testing in four more fields. This is the chest organs CT analysis for the lungs emphysema diagnosis, CT of the chest organs for the diagnosis of pulmonary hypertension signs with the determination of the pulmonary trunk diameter, CT of the brain for stroke diagnosis, X-ray of the musculoskeletal system for flat feet detection.

Neural networks pre-process images and mark up where the doctor's special attention is required. First of all, doctors receive those studies where pathological changes are more likely to be noticeable. Doctors can refer both to the original images and to those images that were processed by artificial intelligence services.

Since last year, an experiment on the computer vision technologies implementation based on artificial intelligence for the medical images analysis has been operating in Moscow. This project is one of the first open city initiatives to attract innovative companies for creating and implementing high-tech services in the social sphere. This was made possible thanks to the unified digital healthcare platform developed by the Moscow Social Development Complex and the Information Technology Department.

In addition, artificial intelligence services were integrated into a unified radiological information service (ERIS EMIAS). They are available to radiologists of all medical institutions connected to the system.


If you continue to use our website, you are agreeing to accept the use of cookies on your device. Cookie files ensure the website’s efficiency and help us provide you with the most interesting and relevant information. Read more about cookie files.
Accept ccokies