The Digital Department of the I.M. Sechenov First Moscow State Medical University, Ministry of Health of the Russian Federation, has created a neural network capable of analyzing electrocardiograms (ECGs) with high accuracy and automatically categorizing them into classes. The model can detect several pathologies of the cardiovascular system in a patient and draw the doctor's attention to this fact.
Effective diagnosis of cardiovascular diseases requires fast and accurate ECG analysis, Gazeta.Ru reports, but due to the large amount of data and lack of qualified specialists in some regions this can be difficult.
The developers claim that automation of the ECG classification process will simplify the work of medical personnel with a large amount of data and help to identify cardiovascular diseases in a timely manner.
ECG classification allows to separate records according to various features, such as rhythm disturbances and ischemia, which will greatly facilitate the work of cardiologists and therapists. The new model is able to detect several pathologies, and the doctor will already confirm or deny their presence, explained the project leader, Alina Kitieva, a student of the digital department of the First Moscow State Medical University.
The use of the new digital system is that the doctor loads the patients' ECG data into the program, which analyzes them and automatically assigns them to classes. The results are transmitted to a specialist for further analysis and decision-making.
The project aims to create a model capable of classifying ECGs with an accuracy of at least 95%. In the future, the development team plans to expand the set of classified heart diseases and implement the system in medical institutions to support medical decision-making.
Earlier in Chelyabinsk, doctors removed a patient's tumor, giving her a chance to become a mother, and chemist Droganova told why shampoo "3 in 1" is easy to spoil hair.