Методы лучевой диагностики все более масштабно применяются при массовых профилактических осмотрах (скринингах) для выявления различных патологических состояний. С целью повышения эффективности программ скрининга в ряде ведущих стран мира предусмотрены двойные описания результатов исследований, что неоспоримо увеличивает рабочую нагрузку на врачей-рентгенологов. В связи с эти крайне актуальной становится задача автоматизированного анализа результатов скрининговых исследований. ЦЕЛЬ ИССЛЕДОВАНИЯ Оценить влияние делегирования полномочий по выполнению первого описания медицинскому программному обеспечению на основе технологий искусственного интеллекта (ИИ) на длительность процесса двойного описания результатов флюорографии. МАТЕРИАЛ И МЕТОДЫ Исследование выполнено на базе Московского референс-центра лучевой диагностики (ГБУЗ «НПКЦ ДиТ ДЗМ»). В исследование включено 13 901 флюорографическое исследование. Реализованы два сценария двойного описания исследований: в первом случае участвовали врач-рентгенолог и алгоритм ИИ, во втором — два врача-рентгенолога. Просмотр результатов ИИ и описание исследований проводилось в Едином радиологическом информационном сервисе Единой медицинской информационно-аналитической системы города Москвы. Выполнен статистический анализ данных. РЕЗУЛЬТАТЫ По сценарию №1 проведено двойное описание 1435 результатов флюорографии, по сценарию №2 — 12 446. В первом сценарии врач, получив данные машинного анализа, затрачивал на подготовку заключения в среднем 0,9±3,0 мин. Во втором сценарии продолжительность работы врача, осуществлявшего первый просмотр, составила 0,8±2,1 мин; второй просмотр — 0,4±1,0 мин. Общая длительность проведения двойного описания в формате «врач + ИИ» колебалась от 0,7 до 1241,2 мин, составив в среднем 199,3±330,3 мин. Во втором сценарии общая длительность проведения двойного описания составила 1 838,5±3671,4 мин. ЗАКЛЮЧЕНИЕ Делегирование первого описания алгоритму искусственного интеллекта принципиально ускоряет предоставление результатов флюорографии, повышая их доступность для обследованных лиц и медицинских работников, направляющих пациентов на обследование. Актуальнейшим вопросом становится точность работы соответствующих технологий искусственного интеллекта, а обязательность их регистрации в качестве медицинского изделия не подлежит дальнейшему обсуждению.

The issues of quality medical care management and organization of the work of the department of radiation diagnostics are always relevant and require constant monitoring and analytical expertise. Since 2018, the Moscow regional branch of the Russian Society of Radiologists and Radiologists (MRO PORR) has been conducting an independent assessment of the departments of radiation diagnostics in all the regions of Russia. The rating aimed to identify industry leaders and spread the best practices throughout the country. The survey results identified the positive trends in the development of diagnostic care services throughout the country and critical points that affect the quality of work of medical organizations. This study presents an analysis of the functioning of 123 departments of radiation diagnostics in 2020. After meeting the inclusion criteria, a list of 163 medical organizations in 15 cities of 7 federal districts was formed. The evaluation procedure was divided into three stages. The first stage consisted of an online survey, wherein each of the participating organizations was asked to answer questions about the departments work arrangement, equipment, list, and features of performing diagnostic tests, as well as working with patients. The second stage consisted of a clinical and technical audit of a set of anonymized studies with conclusions. Special attention was paid to technical audits since several medical organizations violated the methodology of conducting research. The third stage included checking the information about medical organizations in open sources. During the first and second stages, points were awarded, based on which the finalists, leaders, and rating winners were selected. According to the evaluation results of all stages,31 organizations reached the final stage,6 were in the group of leaders, and 5 were winners, whereas 45% of the finalists belonged to the Central Federal District. Greater interest was found in the auditing work in municipal and private medical institutions than in departmental and federal ones. Some database has been collected, in addition to the list of winners, which may represent a cross-section of the state of the radiation diagnostics service in the Russian Federation. Conducting such competitions is primarily aimed at improving the quality and safety of X-ray examinations. The methodology of the competition is improved every year.

BACKGROUND: In Russia, a semi-quantitative CT 04 scoring system is used in the analysis of thoracic computed tomography (CT) scans of COVID-19 patients to grade the severity of lung lesions. Despite the widespread use of this approach, the scoring systems diagnostic accuracy for identification hospitalizations for patients with the disease is currently unknown. AIM: To evaluate the sensitivity, specificity, positive (PPV) and negative (NPV) predictive value of the CT 04 system for the triage of COVID-19 patients. MATERIALS AND METHODS: This retrospective study enrolled 575 patients of Moscow clinics with laboratory-verified COVID-19, aged 57.213.9 years, 55% females. All patients were examined with four consecutive chest CT scans, and the disease severity was assessed using the CT 04 scoring system. Sensitivity and specificity were calculated as conditional probabilities that a patient would experience clinical improvement or deterioration, depending on the preceding CT examination results. For the calculation of the NPV and PPV, we estimated the COVID-19 prevalence in Moscow. The data on total cases of COVID-19 from March 6 to November 28, 2020, were taken from the Rospotrebnadzor website. We used several ARIMA and EST models with different parameters to fit the data and forecast the incidence. RESULTS: The median specificity of the CT 04 scoring system was 69% (95% CI 32%, 100%), and the sensitivity was 92% (95% CI 74%, 100%). The best statistical model describing the epidemiological situation in Moscow was ARIMA (0,2,1). According to our calculations, with the predicted point prevalence of 9.6%, the values of PPV and NPV were 56% and 97%, correspondingly. CONCLUSION: The maximum Youdens index was observed for the period between the first and the second chest CT examinations when the majority of the included patients experienced clinical deterioration. The CT 04 scoring system makes it possible to safely exclude the development of pathological changes in patients with mild and moderate disease (categories CT-0 and CT-1), thereby optimizing the burden on hospitals in an unfavorable epidemic situation.

论证 ：在俄罗斯联邦，为了检测COVID-19肺炎及其并发症和与其他肺部疾病的鉴别诊断，以及对患者进行分类，使用了胸部CT，并在CT 0–4的半定量视觉尺度上评估变化。尽管胸部CT广泛使用，但其用于确定COVID-19患者住院需求的诊断准确性的数字指标目前尚不清楚。 目的 ： 是确定该量表的敏感性、特异性、阳性预测值、阴性预测值。 材料与方法 ：研究涉及575名经实验室确诊的COVID-19患者（55%为女性），年龄为57.2±13.9岁。对于每个患者，进行了4次连续的胸部CT研究，并对疾病的严重程度进行了CT评分（0–4）。根据既往CT研究结果，将敏感性和特异性作为患者病情恶化或改善的条件概率进行计算。为计算阳性预测值（PPV）和阴性预测值（NPV），对COVID-19在莫斯科的流行情况进行了估计。2020年3月6日至11月28日期间所有COVID-19病例的数据来自俄国国家管理的保护消费者服务机构（Rospotrebnadzor）网站。使用了许多具有不同参数的ARIMA和EST模型来选择与现有数据最匹配的模型，并预测发病率的发展。 结果 ：0–4 CT分级的中位特异性为69%，敏感性为92%。描述莫斯科流行病学情况的最佳统计模型是ARIMA（0,2,1）。经计算，预测年发病率为9.6%，PPV值为56，NPV值为97%。 结果 ：Yuden指数最大的阶段出现在胸部CT第一次研究和第二次研究之间，此时样本中大多数患者表现出临床病情恶化的趋势。0–4 CT分级可以安全地排除轻、中度病程（CT0、CT1类）患者的病理变化发展，有助于优化患者在疫情不利的情况下住院。

Telemedicine technologies are successfully implemented and will be used in a global perspective to optimize and improve the quality of diagnostic radiology. On the basis of teleradiology, the concept of centralized radiological reporting has been implemented. The centralization of expertise within the Reference Center provides a complete solution to the issues of staff shortage, uninterrupted research, the maximum level of accessibility and quality of diagnostic radiology. At the same time, the quality and effectiveness of the independent performance of X-ray technicians using telemedicine technologies without the direct presence of a radiologist remain unexplored.The aim of the study. To evaluate the quality and reliability of remote interaction between primary health care X-ray technicians and radiologists of the Reference Center for diagnostic radiology.Materials and methods. The study includes data on the performance of the Reference Center and pilot Moscow city medical facilities providing primary health care (city clinics, n = 12) from January 01, 2019 to March 31, 2021. The number and structure of X-ray technician requests to doctors of the Reference Center for consultations during conducting examinations, as well as a proportion and structure of defects in the performance of X-ray technicians have been relatively studied; satisfaction assessment has been conducted. The following methods were used: analytical, sociological, statistical.Results and discussion. In the conditions of remote performance, the proportion of studies which require communication between X-ray technicians and radiologists is only 0.38 % on average. The share of requests for assistance is decreasing by 35.0 %, while a total number of radiological studies reporting remotely is increasing by 10.0 % monthly. Leading reasons for requests are organizational issues – 45.0 %, admission of an unscheduled patient – 26.0 %. Critical questions on the study methodology are only 11.0 %. Under the working conditions of the Reference Center, the need of X-ray technicians for a doctor’s assistance during the study performance has decreased (p < 0.0001), the pace of getting a consultation if it is required, increased significantly, and time delays have been completely eliminated (p < 0.0001). A satisfaction rate of X-ray technicians with their professional activities has increased (p < 0.0001). In the conditions of remote interaction, the number of technological defects during radiological studies significantly decreased (p < 0.05).Conclusions. A concept of the Reference Center for diagnostic radiology based on the systemic application of telemedicine technologies has been successfully implemented in practice and proven its positive impact on the quality and safety of the performance of primary health care X-ray technicians.

Background and objective: Lung cancer is the most common type of cancer with a high mortality rate. Early detection using medical imaging is critically important for the long-term survival of the patients. Computer-aided diagnosis (CAD) tools can potentially reduce the number of incorrect interpretations of medical image data by radiologists. Datasets with adequate sample size, annotation, and truth are the dominant factors in developing and training effective CAD algorithms. The objective of this study was to produce a practical approach and a tool for the creation of medical image datasets. Methods: The proposed model uses the modified maximum transverse diameter approach to mark a putative lung nodule. The modification involves the possibility to use a set of overlapping spheres of appropriate size to approximate the shape of the nodule. The algorithm embedded in the model also groups the marks made by different readers for the same lesion. We used the data of 536 randomly selected patients of Moscow outpatient clinics to create a dataset of standard-dose chest computed tomography (CT) scans utilizing the double-reading approach with arbitration. Six volunteer radiologists independently produced a report for each scan using the proposed model with the main focus on the detection of lesions with sizes ranging from 3 to 30 mm. After this, an arbitrator reviewed their marks and annotations. Results: The maximum transverse diameter approach outperformed the alternative methods (3D box, ellipsoid, and complete outline construction) in a study of 10,0 0 0 computer-generated tumor models of different shapes in terms of accuracy and speed of nodule shape approximation. The markup and annotation of the CTLungCa-50 0 dataset revealed 72 studies containing no lung nodules. The remaining 464 CT scans contained 3151 lesions marked by at least one radiologist: 56%, 14%, and 29% of the lesions were malignant, benign, and non-nodular, respectively. 2887 lesions have the target size of 3-30 mm. Only 70 nodules were uniformly identified by all the six readers. An increase in the number of independent readers providing CT scans interpretations led to an accuracy increase associated with a decrease in agreement. The dataset markup process took three working weeks. Conclusions: The developed cluster model simplifies the collaborative and crowdsourced creation of image repositories and makes it time-efficient. Our proof-of-concept dataset provides a valuable source of annotated medical imaging data for training CAD algorithms aimed at early detection of lung nodules. The tool and the dataset are publicly available at https://github.com/Center- of- Diagnostics- and-Telemedicine/ FAnTom.git and https://mosmed.ai/en/datasets/ct _ lungcancer _ 500/ , respectively. (c) 2021 Elsevier B.V. All rights reserved.

This dataset contains anonymised human lung computed tomography (CT) scans with COVID-19 related findings, as well as without such findings. A small subset of studies has been annotated with binary pixel masks depicting regions of interests (ground-glass opacifications and consolidations). CT scans were obtained between 1st of March, 2020 and 25th of April, 2020, and provided by municipal hospitals in Moscow, Russia. Permanent link: https://mosmed.ai/datasets/covid19_1110. This dataset is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported (CC BY-NC-ND 3.0) License. Key words: artificial intelligence, COVID-19, machine learning, dataset, CT, chest, imaging

B a c k g r o u n d . In 2019, the Moscow Government decided to conduct a large-scale scientific research – an the Experiment on the use of innovative computer vision technologies for medical image analysis and subsequent applicability in the healthcare system of Moscow (www.mosmed.ai). O b j e c t i v e – analyze engagement, attitudes and feedback from doctors-radiologists in frame of the Experiment. M a t e r i a l s a n d m e t h o d s . The Experiment is a prospective research approved by the Independent Ethics Committee and registered with Clinicaltrails.gov (ID NCT04489992). Patients signed informed voluntary consent. On the date 01.10.2020, ten services are involved in the Experiment, they providing automated analysis of chest computed tomography and x-ray, mammography. The study includes quantitative indicators of the Experiment from 06/18/2020 to 10/01/2020. Methods of social survey, descriptive statistics, assessment of diagnostic accuracy metrics were used. R e s u l t s a n d d i s c u s s i o n . During the first four months of the active phase of the Experiment, ten computer vision services were integrate into Unified Radiology Service of Moscow. More then 497 thousand studies have been successfully analyzed. Analyzes is carried out for 884 diagnostic devices in 293 medical organizations, 272 of them are actively involved. The involvement of medical organizations is 82%. The median time for automatic analysis of 1 study is 8 minutes. Overall, 63% of studies were analyzed in less than 15 minutes. At the beginning of the Experiment, 538 doctors had access to the system; in four months this number increased to 899. The involvement of doctors was 24%, which is slightly higher than the global indicators. According to the results of a sociological survey, the attitude to AI technologies of Moscow radiologists can be characterize as expectant, moderately optimistic. Radiologists have determined that the results of computer vision services are fully consistent with the real situation in 64% of cases. In 36% cases some inconsistencies were recorded; of this number, significant discrepancies took place in 6%, insignificant – in 23%. C o n c l u s i o n . Results of the Experiment’s first four months can be consider as successful. A high level of involvement of radiologists is define. Special measures will be implement to increase the involvement of radiologists, as well as a comprehensive comparative assessment of the work of services at the further stages of the Experiment.

Уже в начале первой волны пандемии COVID-19 для компьютерной томографической (КТ) диагностики поражения лёгких у пациентов с подозрением на вирусную пневмонию в Москве была сформирована сеть амбулаторных КТ-центров (АКТЦ) с круглосуточным режимом работы. Введение шкалы «КТ 0-4» позволило проводить эффективную маршрутизацию. Для предотвращения распространения инфекции среди пациентов и персонала было введено зонирование АКТЦ с разбиением на «красную», «буферную» и «зелёную» зоны. В рамках мобилизации службы лучевой диагностики создан Московский референс-центр, осуществляющий контроль качества, экспертные дистанционные консультации и организационно-методическое сопровождение. Разработано несколько дистанционных курсов и обучающих вебинаров. Для распознавания признаков COVID-19 и оценки степени тяжести были подключены сервисы искусственного интеллекта. Разработанная стратегия службы лучевой диагностики г. Москвы обеспечила готовность к высокой нагрузке на систему здравоохранения города и позволила минимизировать потери среди медицинского персонала. Специалисты службы внесли существенный вклад в эффективное сдерживание распространения инфекции за счёт доступной, своевременной и качественной диагностики и маршрутизации.

Speech recognition technology was tested to prepare protocols for radiological examinations. A method to evaluate the efficiency of speech recognition technology implementation for the preparation of radiological examination protocols has been developed. A time-study was conducted to confirm the effectiveness of voice input. The commitment of radiologist to using innovative technologies in their work practices was evaluated.

New coronavirus infection (COVID-19) viral pneumonia diagnosed by a complex assessment of the epidemiological history, clinical symptoms, radiological and laboratory tests. Radiologists often play a leading role in diagnosis of viral pneumonia, since they may encounter suspicious changes according to radiological studies before clinicians. However, in a number of diseases, including diseases of non-infectious non-viral etiology with a corresponding similar clinical symptoms, it may be difficult to correctly assess the changes detected by computed tomography. This study uses clinical cases to show the main differential diagnostic criteria for COVID-19 viral pneumonia and non-infectious lesions such as pulmonary edema, pulmonary embolism, acute hypersensitive pneumonitis, drug-induced pneumonitis, non-specific interstitial pneumonia, and adenocarcinoma. All patients were hospitalized based on the results of computed tomography, where a diagnosis of non-infectious non-viral lung injuries was established based on morphological and/or typical clinical symptoms, laboratory or radiological data. We examined clinical cases with radiological signs similar to viral pneumonia, such as areas of ground glass opacities with the presence or absence of areas of consolidation, as well as thickening of the lung interstitium with decreased lung attenuation (crazy paving symptom). In a difficult epidemiological situation, it is important for a radiologist to suspect the above-mentioned pathological conditions in patients who are urgently admitted to outpatient CT centers.

At the beginning of the first wave of the COVID-19 pandemic, a network of outpatient CT centers (OCTC) for lung pathology diagnostics in patients with suspected viral pneumonia with the round-the-clock operation was formed in Moscow. The introduction of the CT 0-4 scale allowed for effective routing. To prevent the spread of infection among patients and staff, OCTC zoning was introduced, dividing into red, buffer, and green zones. As part of the mobilization of the Radiology Service, the Moscow Reference Center was established, aimed at quality control, remote expert consultations, and organizational and methodological support. Several online courses and training webinars have been developed. Artificial Intelligence services were connected to recognize the signs of COVID-19 and assess the severity. The developed strategy of the Moscow Radiology Service ensured readiness for the high burden on the city health care system and minimized losses among medical personnel. The experts significantly contributed to effective infection control through accessible, timely, and high-quality diagnostics and routing.

The objective of the study:  to evaluate the applicability of the automated system for detection of chest diseases during a regular mass screening of the population through assessment of universe parameters of diagnostic accuracy. Subjects and methods.  A retrospective diagnostic study was conducted. The inde x-t est (the method being studied) implied distinction and analysis of  X-r ay films using the software based on convolutional neural networks of  U- N ET  type, which were modified and trained for specific purposes. The reference method used was the double revision of the previously classified  X-r ay films by two qualified roentgenologists with work experience of  8-1 0 years. Two depersonalized samplings of digital  X-r ay films were used: Sample 1 ( n =  140), the ratio of the norm and pathology made 50 : 50; Sample 2 ( n =  150), the ratio of the norm and pathology made 95 : 5. Results.  The following parameters were set up for Samples 1 and 2 respectively: sensitivity ‒ 87.2 and 75.0%, specificity ‒ 60.0 and 53.5%, the prognostic value of the positive result ‒ 68.6 and 8.3%, the prognostic value of the negative result ‒ 82.4 and 97.5%, the area under characteristic curve ‒ 0.74 and 0.64. Conclusions.  The index test can be used only for mass regular screening in the population with low pr e-t est chances of pathology, which is confirmed by the prognostic value of the negative result (97.5%). This technology was recommended for the semiautomatic formation of pulmonary tuberculosis risk groups for consequent verification of the results by a roentgenologist.