Purpose Determination of the diagnostic capabilities of metadevices for breast MR examination in women. Material and methods In the study, two types of metadevices for examining the mammary glands were considered – for imaging in a field with magnetic induction of 3 T and 1.5 T. 11 healthy women of reproductive age were examined, magnetic resonance images of T1 (turbo spin echo) and T1 3D gradient echo (GRE) were obtained based on the Dixon method with fat saturation. The images were evaluated by radiologists on a 5-point Likert scale. Results The images obtained using the metadevices were characterized by acceptable and comparable absolute and relative signal-to-noise ratios comparing them to images obtained using a standard coil at the same spatial resolution and with a decrease in input power by an average of 27 times for 3.0 T. At the same time, for 1.5 T, the input power was reduced by a factor of 15.6, and the signal-to-noise ratio was increased by a factor of 2. For image quality criteria in terms of presence/absence of artifacts, the average score for the metadevice was higher than the score for the specialized coil by 3 T. For 1.5 T, this parameter turned out to be the same, which was probably associated with a lower level of artifacts by 1.5 T than by 3 T in general. Discussion Analysis of the collected assessments of independent experts indicates that the diagnostic characteristics of magnetic resonance images of the mammary glands obtained using ceramic-based (for 3 T) and wire-based (for 1.5 T) metadevices are of a good and average level, and are comparable in terms of all criteria with standard approaches. Conclusions The assessment of the quality of the obtained images demonstrates the acceptable quality of imaging and reflects the possibility of their application in clinical practice, taking into account ongoing improvements and optimization of the entire set of pulse sequences for MRI of the mammary glands.

The aim of the study is to evaluate the efficacy of approaches to sampling during periodic quality control of the artificial intelligence (AI) results in biomedical practice. Materials and Methods The approaches to sampling based on point statistical estimation, statistical hypothesis testing, employing ready-made statistical tables, as well as options of the approaches presented in GOST R ISO 2859-1-2007 “Statistical methods. Sampling procedures for inspection by attributes” have been analyzed. We have considered variants of sampling of different sizes for general populations from 1000 to 100,000 studies. The analysis of the approaches to sampling was carried out as part of an experiment on the use of innovative technologies in computer vision for the analysis of medical images and their further application in the healthcare system of Moscow (Russia). Results Ready-made tables have specific statistical input data, which does not make them a universal option for biomedical research. Point statistical estimation helps to calculate a sample based on given statistical parameters with a certain confidence interval. This approach is promising in the case when only a type I error is important for the researcher, and a type II error is not a priority. Using the approach based on statistical hypothesis testing makes it possible to take account of type I and II errors based on the given statistical parameters. The application of GOST R ISO 2859-1-2007 for sampling allows using ready-made values depending on the given statistical parameters. When evaluating the efficacy of the studied approaches, it was found that for our purposes, the optimal number of studies during AI quality control for the analysis of medical images is 80 items. This meets the requirements of representativeness, balance of the risks to the consumer and the AI service provider, as well as optimization of labor costs of employees involved in the process of quality control of the AI results.