The main difference between artificial intelligence (AI) systems and simple automated algorithms is the ability to learn, synthesize and conclude. The AI system is trained on a set of examples, including pictures, characteristics of patients with a certain disease, then it allows to generalize a lot of such examples and get some general functional dependence, which brings in line the patient data and a certain diagnosis. The system can be named intelligent if this synthetizing ability is realized. Although the AI systems are now becoming more understood and accepted by doctors, a deeper understanding of «how it works» is needed. The article provides a detailed review of the application of methods and models of artificial intelligence in the diagnostics of cancer based on the of multimodal instrumental data. The basic concepts of artificial intelligence and directions of its development are presented. From the point of view of data processing, the stages of development of AI systems are identical. The stages of intellectual processing of diagnostic data are considered in the paper. They include the acquisition and use of training databases of oncological diseases, pre-processing of images, segmentation to highlight the studied objects of diagnosis and classification of these objects to determine whether they are malignant or benign. One of the problems limiting the acceptance of AI systems development by the medical community is the imperfection of the explainability of the results obtained by intelligent systems. Authors pay attention to importance of the development of so-called explanatory intelligence, because its absence currently significantly inhibits the introduction and use of intelligent diagnostic systems in medicine. In addition, the purpose of the article is a way to develop the interaction between a radiologists and data scientists.

Purpose. To evaluate the radiological patterns of a new COVID-19 coronavirus infection. Materials and methods. Review of literature sources. Results. COVID-19 causes the acute severe viral pneumonia. Radiological diagnostics of COVID-19 is very important, because CT can be the first study that shows the signs of viral lung lesion, and allows to assess the severity of the lesion and adverse prognostic signs of its further development. The initial CT pattern of COVID-19 is a pattern of infiltration of secondary pulmonary lobules on the type of «frosted glass» (a symptom of «dry leaf») with a subsequent decrease in the volume of lesions at favorable results, or their increase, accession of CT pattern of «cobblestone road» and the appearance in the area of «frosted glass» the alveolar infiltration in unfavorable course of disease. These symptoms are the precursors to the development of respiratory distress syndrome. At a later primary examination, the primary CT symptoms are the pattern of «cobblestone road» and areas of alveolar infiltration, which correlates with an unfavorable further course and outcome. There was noted that viral pneumonia in COVID-19 was characterized by the location of changes in the posterior subpleural and peribronchial areas. All authors confirmed that cavities, nodules, pleural and pericardial effusions, and lymphadenopathy were absent in COVID-19. In the course of observation, quantitative characteristics of the lesions with a score were proposed, the use of which can help in determining the prognosis. Also identified temporary staging of the process and the formation in some of patients the residual changes in the lungs the same as in influenza pneumonia H1N1 (2008–9гг, 2015–16.) and SARS SARS-CoV-2 (2003)which can start the process of development of progressive pulmonary fibrosis. There is a need for frequent CT studies (every 4 days) to enable timely assessment of rapid dynamics and changes in treatment tactics. The analysis of the results of the examination should be performed by at least 2 radiologists experienced in thoracic radiology, with the involvement of a third independent expert, in case of disagreement. All the authors confirmed the low information content of traditional radiography in assessing viral lung lesions. In some studies, chest radiographs were not performed, only CT as a more sensitive method for detecting early changes, similar to previous outbreaks of coronavirus. However, the role of traditional radiography was recognized as unquestionable when evaluating changes in reanimation department conditions. Conclusions. The accumulation of experience in clinical and radiological examination of COVID-19 patients allowed to determine the radiological semiotics of the process, which is important for determining the treatment tactics.

The patients with gastric cancer (GC) sometimes have a very poor prognosis. In general, it depends on the time of GC verification, which dramatically changes prognosis and treatment plans. This article reflects current data on the epidemiology, mortality rate, and rates of the incidence of gastric cancer in different countries of the world. The most common and clinically convenient classifications and staging systems for gastric cancer are given, the need for the earliest possible diagnosis and verification of the stage of the tumor process is substantiated. There are considered in details the modern methods of the gastric cancer diagnostics, including esophagogastroduodenoscopy (EGD), as the method of choice for the primary detection of gastric cancer, and various methods of radiation imaging, including multispiral computed tomography and endoscopic ultrasonography (EUS), combined with chromoendoscopy, narrow-band tomography, and confocal laser endoscopy. Particular attention is paid to the presentation of various methods of preoperative T-staging and N-staging of gastric cancer, comparing the diagnostic accuracy of MSCT and endoscopic ultrasonography. Diagnostic accuracy, sensitivity, and specificity of magnetic resonance imaging (MRI) and positron emission tomography (PET) are presented. Special attention is paid to the comparative analysis of the effectiveness of various imaging techniques in assessing the effectiveness of neoadjuvant chemotherapy; the results of tumor volumetry are compared with the histological response according to the Mandard score. Further prospects for the application and development of radiological research techniques are discussed.

Introduction. Neurodegenerative diseases occupy a significant place in the structure of diseases of the Central nervous system and have a high social and economic significance. One of the methods of more accurate neuroimaging assessment of the neurodegenerative process is magnetic resonance morphometry using specialized software. Purpose of research. To evaluate the possibilities of magnetic resonance morphometry as a method of differential diagnosis of parkinsonism syndromes and to investigate the characteristic of volumetric changes of individual brain structures in different variants of Multisystem atrophy. Materials and methods. Morphometric markers of Multisystem atrophy variants development using postprocessing processing of native MRI on Freesurfer software are analyzed. 24 patients with striatonigral degeneration and 20 patients with olivopontocerebellar degeneration were examined. Patients without relevant neurological pathology of appropriate age were examined as a control group. Magnetic resonance imaging was performed on a magnetic resonance tomograph of Siemens with magnetic field induction of 1,5 T. 240 structures from the right and left hemispheres of the brain were subjected to statistical analysis. There were significant differences in the distribution of atrophy among the structures of the brain stem and hemispheres in patients with olivopontocerebellar and striatonigral degeneration. Сonclusions. The obtained data of the study on the atrophy of cortical areas involved in the realization of cognitive functions showed the difference between the variants of Multisystem atrophy. Thus, striatonigral atrophy is characterized by a decrease in the thickness of the structures of the frontal cortex, while olivopontocerebellar degeneration reveals a decrease in the thickness of the parahippocampal zone. Methods of postprocessing processing of MR morphometry data allow to confirm the clinical diagnosis instrumentally, as well as to clarify the pathogenesis of neurodegenerative diseases and to clarify the distribution of the brain structures most susceptible to neurodegeneration.

The introduction of antiretroviral therapy has changed the human immunodeficiency virus pandemic. Some patients with HIV infection after starting or resuming ART develop a paradoxical worsening of clinical status, called Immune Reсоnstitution Inflammatory Syndrome (IRIS). However, if clinical and laboratory criteria for the diagnosis of this syndrome have been formulated, IRIS neuroradiological criteria do not exist yet. The present study presents neuroradiological features and diagnostic algorithm for identification of IRIS involving central nervous system.

To analyze the information content of automatic recognition of formations in lungs at digital radiography on the example of one of the commonly available diagnostic algorithms. Materials and methods. This study is based on the results of recognition and analysis of digital radiographs from test bases by software product, based on convolutional neural networks on the example of one of the generally available diagnostic algorithms. The database consisted of anonymized digital radiographs of 240 patients whose health condition was confirmed by histological data and computed tomography (CT) results, the ratio between normal and pathology was 50:50%. Results. At application of automatic recognition of formations in the lungs on digital radiographs, good possibilities were revealed for detecting rounded formations in the lung tissue (93% of pathological changes). Recognition results with a solid structure of formations were higher than with a subsolid — 94% and 88%, respectively. A significant dependence of the possibility of detecting formations in the lungs on their size was revealed. For the focal lesions smaller than 10 mm in none of the cases the algorithm determined them with a probability of more than 50%, and only in 33% of cases the probability was 10–50%. Indicators are significantly higher at the size of formations from 10 to 305 mm (61% of solid formations and 25% of subsolid formations with a probability indicator more than 90%). When the size of solid formations is more than 30 mm, all the 100% of cases are revealed, while in 73% the probability is indicated more than 90%. There was also identified 100% of the subsolid formations. Сonclusion. The application of automated systems as a method of interpretation of radiographs is a promising direction, requiring further improvement and more detailed study of the results obtained. The diagnostic effectiveness of these algorithms is close to effectiveness of radiologists. Currently, the results obtained by the algorithm as a software product for identifying of rounded formations in the lungs cannot be used as a reliable diagnostic method, but it might be considered as an auxiliary «second reading» for the radiologist.

Introduction. The direct relationship between the level of sodium intake, arterial hypertension, followed by the development of heart failure, a hypothesis of the direct influence of excessive sodium accumulation in myocardial glycosaminoglycans seems quite probable, which can further contribute to the occurrence of diastolic dysfunction and heart failure.The aim of the study was to identify excess sodium in rats under conditions of sodium loading, in comparison with rats at a normal level of sodium intake. Materials and methods. Ten male Wistar rats with the same body weight were divided into two groups: the excess salt intake group and the normal salt intake group. Estimation of the amount of Na and NaCl in the animal myocardium was performed using dual energy computed tomography (DECT) samples. Samples were scanned on a Revolution GSI tomograph (GE Healthcare). For statistical processing of the obtained data, the R language was used. Results. The results of the study showed that the accumulation of Na and NaCl does not depend on the average level of animal feed intake, there is no correlation between weight and accumulation of excess Na in tissues, the level of Na and NaCl detected in myocardial tissue significantly increases the likelihood of a high salt diet in rats, and a relationship between the content Na in the myocardium and NaCl+H₂O. Conclusion. The experiment confirmed the existence of a reliable relationship between the sodium compounds calculated on the basis of DECT and the content of these compounds in the samples. The small number of samples did not allow us to calculate normalized rats, but we noted a clear difference between the control group and the high sodium diet.

Transthoracic core-needle biopsy is a minimally invasive and effective procedure for histological verification of intrathoracic lesions. It’s accuracy can be assisted by ultrasound or robotic navigation. Material and methods. Retrospective analysis of 45 transthoracic core-needle biopsies under ultrasound and robotic guidance. Results. The accuracy of both methods is over 90% and complications rate is comparable. Ultrasound guidance is more suitable for large subpleural lesions, attached to the chest wall. Robotic guidance allows more precise positioning for small lesions, but takes much more time. Complications rate for ultrasound-guidance was 5,3% vs 15,4% for robotic guidance. However, none of these complications required surgical intervention. Even 3 tissue samples provided enough material for immunostaining and for molecular analysis if necessary. Summary. Transthoracic core needle biopsy is a method of choice for histological verification of large solid intrathoracic lesions, adherent to the chest wall, without large cavities of destruction. Such patients have the lowest risk rate of possible complications and are suitable for multiple biopsies. At the same time, these patients do not fit for VATS due to large size of the tumor and pleural adhesions, which can be difficult to manage during thoracoscopy. Ultrasound guidance can be best of all applied for large solid tumors, invading the chest wall.

Objective: to assess the state of the liver parenchyma in patients with fat overweight using the quantitative ultrasound steatometry method. Material and methods: 118 patients aged 19–75 years were examined, 52 of them men (44%) and 66 women (56%) due to a single 7 (8) stages diagnostic algorithm: questioning and complaints collecting, quality of life, clinical inspection, non-invasive bioimpedancemetry with subsequent assessment of fat mass index, ultrasound liver examination in B-mode, quantitative ultrasound steatometry with ultrasound wave attenuation coefficient evaluation, dual-energy X-ray absorptiometry in the «Whole body» mode, liver biopsy with histological microscopic specimens on SAF and NAS scales assessment. The control group consisted of 46 patients with a normal ration of fat mass and without ultrasound signs of fatty liver infiltration. Results: there were no signs of liver steatosis in 4 patients (3,4%) with a clinical picture of metabolic syndrome were detected using instrumental diagnostic methods. The sensitivity and specificity of B-mode were 60,4% and 72,8%, respectively, quantitative ultrasound steatometry — 90,7% and 92,4%, dual-energy X-ray absorptiometry in the «Whole body» mode — 88,7% and 90,1%. Conclusion: Ultrasound steatometry is an informative method for non-alcoholic fatty liver disease screening. The correlation of quantitative ultrasound diagnostics data for steatosis with a biopsy at stage S0 corresponds to r=0,87; at stage S1 r=0,69, S2 r=0,75, at stage S3 corresponds to r=0,86, which indicates the high informativeness of this method; Dual-energy X-ray absorptiometry in the «Whole body» mode can be used as an alternative to multislice computed tomography to identify components of the metabolic syndrome due to low radiation exposure.

Aim of the study. We have studied the possibility of imaging of breast cancer in dogs by MRI with paramagnetic contrast enhancement using newly proposed agent Mn-dimercaptosuccinate (Mn-DMSA₂, ^TMSuccimang), in dogs. Material and methods. In seven animals (veterinary patients) with breast cancer the MRI study was carried out, when sleeping (infusion of propofol), with contrast enhancement using Mn-DMSA₂, as 0,5M solution, 0,1 mM/Kg of BW. The T1-w MRI scans were acquired before, and in 12–17 min after injection of Mn-DMSA₂, TR=450–600 ms, TE=12–15 ms, matrix 256×256, field of view 250 mm, slice 2,5 mm thick. All studies were carried out using Toshiba Titan Vantage (1,5 Т, five animals) and Siemens Magnetom Open (0,2 Т, in two). The uptake of Mn-DMSA₂ to tumor was quantified using Index of Enhancement, as ratio of intensity per voxel of the post-contrast image to the intensity of the pre-contrast initial one. Results. In all cases the uptake of Mn-DMSA₂ was seen in the areas of both primary tumor and of metastases to lymph nodes and to the brain. Index of enhancement was: over primary tumor (n=7) — 1,52±0,19 (1,35; 1,83), over metastases to lymph nodes (n=12) 1,37±0,14 (1,12; 1,64), over peripheral zone of brain metastases (n=11) — 1,48±0,15 (1,29; 2,37), and over central zone of brain metastases (n=11) — 2,76±0,12 (1,48; 3,21). No any case demonstrated any complication or concomitant effect. Conclusion. Mn-DMSA₂ is an perspective non-gadolinium paramagnetic contrast agent useful for diagnosis of both primary tumor and of metastases in breast cancer.

Introduction. Positron emission tomography/computed tomography (PET/CT) is a recommended technique for tumor response evaluation in lymphoma after treatment. The possibilities of the whole body magnetic resonance imaging with diffusion-weighted imaging (MRI-DWI) for tumor response evaluation in lymphoma are not well studied. Objective: to compare the diagnostic effectiveness of whole body MRI with a diffusion-weighted imaging (MRI-DWI) and PET/CT in determining tumor response in lymphoma after the end of chemotherapy. Materials and methods. A prospective study included 105 adult patients with lymphoma whounderwent whole body MRI-DWI and PET/CTafter the end of chemotherapy and who were followed-up for at least 6 months. To interpret the MRI-DWI, the 5-level scale (Minsk scale) proposed by us was used. Categories 1–3 were considered a sign of complete tumor response, categories 4–5 were a sign of non-complete response. Results. According to the reference standard, complete tumor response was established in 77% of patients, non-complete response in 23%. The assessment of the tumor response in MRI-DWI and PET/CT matched in 89% of patients. The agreement of MRI-DWI (k =0,76, p=0,000) and PET/CT (k =0,78, p=0,000) with the reference standard is good. The sensitivity, specificity, accuracy, positive and negative prognostic value of MRI-DWI were 66,7%, 100,0%, 92,4%, 100,0%, 91,0%, PET/CT — 83,3%, 95,1%, 92,4%, 83,3%, 95,1%, respectively. The diagnostic effectiveness of the methods is not significantly different (p=0,32). The most common reason for the incorrect determination of the tumor response in MRI-DWI was non-enlarged lymph nodes, and in PET/CT — metabolically active non-tumor diseases. 3-year progression-free survival with negative and positive MRI-DWI results was 93% and 25% (p=0,000), 3-year overall survival — 97% and 70% (p=0,011), respectively. Conclusion. Whole body MRI-DWI and Minsk scale are recommended for use in patients with lymphoma to determine tumor response after the end of chemotherapy as a non-irradiative and effective alternative to PET/CT.

Introduction. Dysferlinopathy is a phenotypically heterogeneous group of hereditary muscular dystrophies caused by mutations in the dysferlin gene (DYSF). Debut in adolescence, predominantly in physically developed patients, combined with the often subacute development of hypercreatine phosphatemia and edematous muscle changes in MRI often leads to diagnostic errors. Purpose of the study: to determine the most typical MRI pattern of muscle damage of the pelvic girdle and lower limb in patients with dysferlinopathy. Materials and methods. 40 people were examined, among which 20 patients with a clinical picture of dysferlinopathy with an equal ratio of Miyoshi phenotypes and LGMD and an average age of 35 (24; 44) years. Comprehensive clinical and instrumental examination included neurological, electroneuromyographic and molecular genetic studies (NGS). Magnetic resonance imaging of the muscles of the pelvic girdle and lower limb was performed in 20 patients and a control group equivalent in sex and age. Results. The use of semi-quantitative MRI indicators — relative signal intensity — D (D=T1 muscle (STIR) / T1 (STIR) subcutaneous fat layer) made it possible to formulate the characteristics of a common typical MRI pattern of muscle involvement in dysferlinopathy. An increase in the intensity of the relative signal D, T1 in the rear muscle group of the thighs and lower legs, indicating fatty infiltration was observed most frequently, while a decrease in D, STIR values was observed in the anterior and medial muscles of the thighs, reflecting the presence of edema of the previous fatty degeneration of these muscles. Conclusion. In addition to the general idea of muscle involvement in dysferlinopathy, it is advisable to consider the «early», «typical / completed» and «late» MRI patterns of dysferlinopathy that increase the effectiveness of the diagnosis of this disease. In the differential diagnosis of the Miyoshi phenotype from LGMD, one should focus on maintaining normal values of D, T1 from m. gluteus maximus and m. popliteus at all stages of the disease.

British Society of Thoracic British Society of Thoracic British Society of Thoracic Imaging.