ABAleksander Vladimirovich Bazhin

Aleksander Vladimirovich Bazhin

*This is an algorithmically generated author record based on publicly available information.

Aleksander Vladimirovich Bazhin has a total of 72 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Using Bone-RADS to determine treatment strategies for incidental solitary bone lesions on CT and MRI in adults: a case series

    by Andrey Alekseevich Uchevatkin, Учеваткин Андрей Алексеевич, Anastasia Sergeevna Nered, Неред Анастасия Сергеевна, Aleksander Vladimirovich Bazhin, Бажин Александр Владимирович, Elmar Rasimovich Musaev, Мусаев Эльмар Расимович, Irena Vladislavovna Boulytcheva, Булычева Ирина Владиславовна

    Detecting and differentially diagnosing bone tumors and tumor-like lesions remain major radiological challenges due to variability of pathological processes and corresponding radiographic features, which can be similar across different pathological entities. Following 1993, when the American College of Radiology had introduced the Breast Imaging Reporting and Data System (BI-RADS), a system that later became widely adopted in radiology, many similar systems emerged, the most common of which are the Prostate Imaging Reporting and Data System, the Liver Imaging Reporting and Data System, and the Thyroid Imaging Reporting and Data System. Since 2021, non-Russian colleagues have proposed several frameworks to guide the assessment of focal bone lesions and help determine treatment strategies, including the Solitary Bone Tumor Imaging Reporting and Data System, the Osseous Tumor Reporting and Data System, and the American College of Radiology Bone Reporting and Data System. In September 2022, the Standards and Guidelines Committee of the Society of Skeletal Radiology, drawing from a literature review and expert consensus, introduced a Bone Reporting and Data System for grading incidental solitary bone lesions on computed tomography and magnetic resonance imaging. This system categorizes findings into four management tiers that guide patient management: Bone-RADS 1 (likely benign: leave alone); Bone-RADS 2 (incompletely assessed on imaging: additional imaging modality required); Bone-RADS 3 (follow-up imaging); Bone-RADS 4 (oncologic referral and/or biopsy necessary). Evaluation of lesion CT density (lucent vs sclerotic/mixed) and its MRI signal characteristics allow to assign a case-specific Bone-RADS category and determine suitable approach to patient management. This clinical case series illustrates our experience applying these algorithms proposed by non-Russian colleagues to guide patient care. Case reports incorporate translated content adapted from the open-access publication [DOI: 10.1007/s00256-022-04022-8], published in the journal Skeletal Radiology in September 2022 under the CC-BY 4.0 open license.

  • Open Access

    Transformation of the interaction between the orthopedic traumatologist and radiologist in a trauma center setting: the potential of artificial intelligence

    by Anton V. Vladzymyrskyy, Владзимирский Антон Вячеславович, Alexander V. Bazhin, Бажин Александр Владимирович, Dina A. Akhmedzyanova, Ахмедзянова Дина Альфредовна, Yuliya F. Shumskaya, Шумская Юлия Федоровна

    BACKGROUND: Traumatologists often order and independently interpret radiology studies in outpatient trauma care, since radiologists’ reports may be delayed by several hours, rendering them largely formal. Artificial intelligence (AI) could assist traumatologists in real-time image analysis to prevent diagnostic errors and improve quality of care. AIM: To evaluate the possibility of using AI technologies for fracture detection in a trauma outpatient setting. METHODS: A multicenter, retrospective, observational study was conducted. Three AI tools were assessed. For the shoulder, wrist, and ankle joints, a dataset of 100 anonymized X-rays was compiled, annotated by three radiologists. AI outputs were compared against expert annotations using contingency tables. Diagnostic accuracy was evaluated via receiver operating characteristic analysis. RESULTS: For shoulder X-rays, AI demonstrated an accuracy of 0.95 (95% CI: 0.91–0.99), sensitivity of 0.92 (95% CI: 0.84–1.0), and AUROC of 0.96 (95% CI: 0.92–1.00). Wrist fracture analysis showed accuracy of 0.93 (95% CI: 0.88–0.98), sensitivity of 0.9 (95% CI: 0.82–0.98), and AUROC of 0.95 (95% CI: 0.90–0.99). Ankle studies yielded accuracy of 0.89 (95% CI: 0.83–0.95), sensitivity of 0.82 (95% CI: 0.71–0.93), and AUROC of 0.90 (95% CI: 0.84–0.96). CONCLUSIONS: The current level of AI technology development allows the use of AI-based software in emergency outpatient trauma care, thereby optimizing care processes in the trauma center and eliminating radiologist reporting of X-ray results.

  • The quality of automated detection of sinusitis based on radiography results in practical healthcare (prospective diagnostic study)

    by A. V. Bazhin, A. V. Vladzymyrskyy, K. M. Arzamasov, Yu. A. Vasilev, S. S. Semenov

    Purpose. To study the diagnostic accuracy of software based on artificial intelligence technologies for the diagnosis of maxillary and frontal sinusitis (J01, J32) when used in practical healthcare. Materials and methods. Prospective diagnostic study (10.2023–12.2024) using the STARD 2015 methodology. The Index Test is four anonymized AI services for analyzing the X-ray imaging of the paranasal sinuses, integrated into Unified Radiological Information Service (ERIS) as part of the Moscow Experiment. The Reference Standard is an original methodology of clinical monitoring with a double expert review, an assessment of accuracy according to the criteria of “interpretation” and “localization”, calculation of sensitivity, specificity and AUC. After testing in the ERIS circuit, the AI services worked with real data and underwent monthly clinical monitoring. Results. From October 2023 to December 2024, AI services analyzed 126,547 radiographs of the paranasal sinuses, of which 53% were processed by AI service-4. The scope of the research varied due to the different duration of the services. Regular clinical monitoring covered 2,320 radiographs. The correctness of pathology detection, the accuracy of its classification and localization were evaluated. The highest diagnostic accuracy was found in AI services-1 and 2 (AUC 0.98 and 0.95; integral score – 95.1% and 83.3%). They demonstrated high specificity and precise localization of pathologies. AI services-3 and 4 demonstrated comparatively lower results (AUC 0.89; integral score – 76.6% and 79.4%), with lower specificity for AI service-3 (0.65). Conclusions. As part of the Moscow Experiment, AI services successfully analyzed 126,547 radiographs of the paranasal sinuses (2023–2024), showing an average level of accuracy (AUC 0.93) and quality (83.6%). The results obtained confirm the applicability of AI in clinical practice. Differences in accuracy between services require regular monitoring of safety and quality. The developed methodologies make it possible to effectively carry out such control. The achieved level of quality opens prospects for the transformation of the interaction of doctors of different specialties, which will be the topic of future research.

  • The feasibility of double automated reading of chest radiographic screening results (based on the Moscow experiment on computer vision in radiology)

    by A. V. Bazhin, Y. A. Vasilev, A. V. Vladzymyrskyy, K. M. Arzamasov, I. M. Shulkin

    Introduction: Mass preventive chest imaging examinations (fluorography and X-ray) play a critical role in identifying diseases of public health significance, yet place an additional burden on the healthcare system. Automated sorting of imaging studies using medical devices (MD) powered by artificial intelligence (AI) technologies facilitates optimization by relieving radiologists from having to review studies without pathological findings. However, despite high AI accuracy, rare false-negative findings remain possible, which can be critical in medical screening. Objective: The purpose of this retrospective study was to evaluate an automated dual reading method for autonomous sorting of preventive imaging studies. Materials and methods: The study included 411 clinical cases previously misclassified as «normal.» Each study was reanalyzed by a different AI model. Results: As a result, the re-classification yielded 31.6% correct pathology identification and reduced clinically important discrepancies by 25.5%. When simulating the automated dual reading, the proportion of false-negative findings decreased from 0.071% to 0.052%. Discussion: The simulation revealed significant performance variability across the AI models, emphasizing the need for careful post-authorization monitoring to replace underperforming applications. Conclusion: Thus, automated dual reading improves the safety of autonomous sorting by minimizing the number of missed pathological findings. At the same time, the method requires a strategic approach to AI model selection as it poses a risk of reducing the system’s operating efficiency. The optimal benefit-risk ratio should be determined against contribution of preventive programs to public health interest.


  • Open Access

    READINESS OF ORTHOPEDIC TRAUMA SURGEONS TO ADOPT ARTIFICIAL INTELLIGENCE TECHNOLOGIES FOR THE ANALYSIS OF RADIOLOGICAL EXAMINATIONS: A CROSS-SECTIONAL STUDY

    by Alexander V. Bazhin, Бажин Александр Владимирович, Georgii A. Airapetov, Айрапетов Георгий Александрович, Anton V. Vladzymyrskyy, Владзимирский Антон Вячеславович, Yuliya F. Shumskaya, Шумская Юлия Федоровна, Dina A. Akhmedzyanova, Ахмедзянова Дина Альфредовна

    BACKGROUND. In clinical practice, orthopedic trauma surgeons are often required to interpret diagnostic images without the immediate availability of official radiology reports. The use of artificial intelligence (AI) has the potential to enhance diagnostic accuracy and the overall quality of care. However, the success of AI implementation depends on its acceptance by doctors. Sociological surveys allow assessing awareness of AI capabilities, readiness for its adoption, and the key barriers to their integrating. AIM. To evaluate the attitudes of orthopedic trauma surgeons toward AI, with a specific focus on its application for the autonomous interpretation of radiological studies. METHODS. A cross-sectional anonymous online survey was conducted among orthopedic trauma surgeons working in Moscow healthcare institutions. Data was collected using an adapted version of the validated ATRAI-14 questionnaire, designed to assess radiologists’ attitudes toward AI. Data were analyzed using descriptive statistics. RESULTS. The majority of orthopedic trauma surgeons (62.8%) reported no prior experience with AI for radiological interpretation, and 41.9% indicated a lack of access to such tools in their current practice. Nevertheless, 39.5% expressed willingness to use AI for analyzing radiological studies, while only 18.6% reported actual use of it. The most highly valued functionality, cited by more than one-third of respondents (32.6%), was autonomous triage of studies into “normal” versus “pathological,” followed by specialist interpretation of identified abnormalities supported by decision-support systems. CONCLUSIONS. Orthopedic trauma surgeons’ attitudes toward AI are neutral and skeptical, reflecting the early stage of innovation adoption in clinical practice. To shift this perception, it is essential to build a robust evidence base for the effectiveness of AI and to implement educational initiatives aimed to develop relevant competencies among specialists.

  • Open Access

    Errors in automated triage of preventive chest scans: main causes and potential reduction strategies

    by A. V. Bazhin, Yu. A. Vasilev, A. V. Vladzymyrskyy

    Introduction. The high prevalence of chest X-ray and fluorographic examinations creates significant workload for the radiology department. One promising solution for optimizing the preventive radiography and fluorography pipelines is employing artificial intelligence (AI) technologies. This approach demonstrates effectiveness in triaging studies based on binary assessment (“normal”/”abnormal”). Evaluation of this method identified shortcomings in the form of false-negative results requiring further study and improvement.Aim: To evaluate potential methods for reducing false-negative results in automated triage of preventive chest radiographic examinations using AI-powered software.Material and Methods. In a retrospective diagnostic study conducted from May 1, 2025, to June 30, 2025, we analyzed 411 cases of false-negative results identified during evaluation of automated triage powered by three different AI models. The study utilized data from preventive chest radiographic examinations performed in outpatient setting and mass screening examinations of the adult population in Moscow. Images were re-analyzed by two expert radiologists (> 10 years of experience) considering the available clinical information and patient history data. Analytical and descriptive statistical methods were used.Results. The majority (n = 266) of false-negative results were associated with missed infiltrative changes and pulmonary foci. The remaining clinically significant discrepancies were observed in 24 cases.Analysis of archival data and patient histories identified additional information that, when integrated into AI-powered services, could influence study interpretation and reduce false-negative risk. Such data were found in 68.3% of cases (n = 198) with clinically significant omissions and in 5% (n = 6) with clinically insignificant omissions.Conclusion. Reducing false-negative results requires an integrated approach: implementing duplicate AI services, improving physician referral quality, and integrating clinical data and previous patient study results into AI algorithms, necessitating development of multimodal AI software.

  • Open Access

    Autonomous artificial intelligence for sorting results of preventive radiological examinations of chest organs: medical and economic efficiency

    by Yuriy A. Vasilev, Dmitry A. Sychev, Alexander V. Bazhin, Igor M. Shulkin, Anton V. Vladzymyrskyy, Alexandra Yu. Golikova, Kirill M. Arzamasov, Andrei V. Mishchenko, Gevorg A. Bekdzhanyan, Arcadiy S. Goldberg, et al.

    BACKGROUND: This article proposes a model for organizing preventive radiological examinations of chest organs through autonomous sorting of examination results using medical devices based on artificial intelligence technologies, optimized for maximum sensitivity — 1.0 (95% CI: 1.0; 1.0). Sorting involves classifying the results of mass preventive screenings (fluoroscopy and chest X-rays) into two: “not normal” and “normal.” The “not normal” category includes all cases of abnormalities (e.g., pathological conditions, post-disease or post-surgery consequences, and age-related and congenital features), which are sent for interpretation by a radiologist. The “normal” category consists of cases without signs of pathological deviations, which potentially do not require a radiologist’s description. AIM: To evaluate the feasibility, effectiveness, and efficiency of autonomous sorting of results from preventive radiological examinations of chest organs. MATERIALS AND METHODS: A prospective multicenter diagnostic study was conducted on the safety and quality of autonomous sorting of results from preventive radiological examinations of chest organs. Analytical and statistical methods of scientific inquiry were used. RESULTS: The study included results from 575,549 preventive radiological examinations obtained through fluoroscopy and chest X-rays and processed using three medical devices based on artificial intelligence technologies. In autonomous sorting, 54.8% of the preventive radiological examinations of chest organs were classified as “normal,” resulting in a proportional reduction in the radiologist’s workload for interpreting and describing the examination results. Fully correct autonomous sorting was achieved in 99.95% of cases. Clinically significant discrepancies were determined in 0.05% of cases (95% CI: 0.04; 0.06%). CONCLUSIONS: This study confirmed the medical and economic effectiveness of the model for autonomous sorting of results from preventive radiological examinations of chest organs using medical devices based on artificial intelligence technologies. The next phase should involve updating the regulatory framework and ensuring the legitimacy of the autonomous application of certain medical devices based on artificial intelligence technologies in established conditions and preventive tasks.


  • Open Access

    Optimization of magnetic resonance imaging of the hand

    by Yuriy A. Vasilev, Dmitriy S. Semenov, Alexey V. Petraikin, Andrey A. Uchevatkin, Liya R. Abuladze, Alexander V. Bazhin, Dariya E. Sharova, Петряйкин Алексей Владимирович, Practical Clinical Center For Diagnostics, Telemedicine

    BACKGROUND: Magnetic resonance imaging is one of the leading imaging modalities of the musculoskeletal system. However, when imaging the hand, major problems in magnetic resonance imaging include the lack of specialized coils and reliable fixation devices for the hand, uncomfortable patient posture, motion artifacts, and small anatomical structures in the wrist. These factors inevitably lead to incorrect interpretation. AIM: To improve the quality of magnetic resonance imaging of the hand by developing an approach to coil selection, scanning protocol, and hand positioning and fixation. MATERIALS AND METHODS: A positioning device was developed to prevent hand movements. Two types of coils were evaluated. Magnetic resonance images were evaluated comparatively, as well as by a musculoskeletal radiologist. RESULTS: А head coil is more appropriate when scanning the entire hand, for example, in rheumatic diseases. A knee coil is more appropriate when studying smaller anatomical structures (including the wrist) owing to a smaller field of view and higher resolution. Based on the obtained data, guidelines for the selection of scanning parameters, sequences, and coils for magnetic resonance imaging of the hand were formulated. To prevent motion artifacts, a special fixation device of the patient’s hand was introduced. CONCLUSION: Certain factors directly affect the qualitative magnetic resonance imaging study of the hand, such as safety protocols, scanning parameters, and hand fixation. The guidelines presented in this study and the use of the developed specialized fixation device may improve the quality of magnetic resonance imaging of the hand.

  • Open Access

    Magnetic resonance imaging of the temporomandibular joint in patients with metal structures

    by Yu. A. Vasilev, D. V. Burenchev, Yu. N. Vasileva, O. Yu. Panina, V. A. Ignatieva, D. V. Dushkova, A. V. Bazhin, E. G. Privalova, V. A. Ulyanova

    High-quality visualization of the temporomandibular joint (TMJ) allows to choose the most optimal tactics for conservative or surgical treatment. Performing magnetic resonance imaging (MRI) in patients with metal structures in the maxillofacial area is a pressing issue in ensuring quality diagnostics. In the presented clinical case, a methodology of performing diagnostically significant MRI of the TMJ in the presence of metal structures in the region of interest (ROI) was clearly demonstrated and described.

  • Open Access

    Dynamic MRI in a COVID-19 patient: a case series

    by Evgeniia A. Grik, Olga Yu. Panina, Anna N. Khoruzhaya, Dmitriy S. Semenov, Alexander V. Bazhin, Yulia N. Vasileva

    Extensive spread of the coronavirus disease (COVID-19) prompted an investigation of its diagnostic features. Acute viral pneumonia associated with COVID-19 has been described in detail using CT, radiography, and MRI. There is no data in the literature on the descriptive picture observed with dynamic MRI. Considering a comprehensive diagnostic approach, radiologists should know how to correctly recognize and interpret COVID-19 on MRI. This case series demonstrated the ability of dynamic MRI to detect the cloudy sky sign and distinguish it from consolidation in COVID-19 patients, thus presumably distinguishing between early or mild changes and a progressive clinical course. These changes in dynamic lung images on MRI can be recorded depending on the phase of the respiratory cycle. Thus, MRI, as a radiation-free tool that can be used to examine a patient with acute viral pneumonia COVID-19, can be useful in cases where access to computed tomography is limited and dynamic morphofunctional imaging is required.