EKElena Kleymenova

Elena Kleymenova

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

Elena Kleymenova has a total of 43 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Speech recognition applications for surgical safety checklist verification: a systematic review

    by Nodari M. Kakabadze, Какабадзе Нодари Малхазович, Elena B. Kleymenova, Клеймёнова Елена Борисовна, Anton G. Nazarenko, Назаренко Антон Герасимович, Luibov Yashina, Яшина Любовь Петровна

    Surgical checklists (SCs) are designed to reduce medical errors and improve staff coordination and accountability. However, the effectiveness of SCs depends on strict adherence to the rules for their use. Artificial intelligence-based voice assistants (VAs) may improve patient safety during surgical verification procedures, but the evidence is fragmentary. The aim of this systematic review was to examine VAs designed for audio recording in the operating room, checklist item verification, and interactive engagement with the surgical team. Of 102 references identified, 44 publications were selected for full-text review; of these, four met the inclusion criteria: one was a clinical trial comparing VA use with manual SC completion, and the remaining three were pilot studies. The risk of bias for non-randomized studies, assessed using ROBINS-I, was serious for the comparative study and moderate for the descriptive studies. All studies used different automatic speech recognition (ASR) models; one also used speaker verification and natural language processing. Speech recognition accuracy, reported in three studies, ranged from 90.7% to 97.3% for ASR; 94.1% for speaker verification; and 92.8% to 94.8% for formalized response and checklist rule verification. VA usability was assessed in one study using the System Usability Scale, with a score of 76.04 ± 11.61. In the comparative study, user satisfaction with the VA was higher than with manual SC completion (p 0.05). In the same study, the VA improved the rate of full SC completion by 16.4% (from 81.8% to 92.5%); the odds ratio for full SC completion was 2.74 (95% CI: 1.55–4.82; p = 0.00032). No study reported adverse event detection using VAs. Thus, all analyzed studies confirmed the feasibility of voice-based SC completion with adequate accuracy and intelligent support for surgical verification. However, there is a lack of high-quality studies confirming the efficacy of VAs in ensuring patient safety during surgery, detecting and preventing adverse events, and identifying deviations from the surgical process.

  • Open Access

    Machine learning methods for recognizing surgical site infection in trauma and orthopedic patients: a cross-sectional study

    by Anton G. Nazarenko, Elena B. Kleymenova, Alexey I. Molodchenkov, Dmitry S. Gorbatyuk, Azat D. Enikeev, Valery A. Kislyakov, Liubov P. Yashina, Горбатюк Дмитрий Сергеевич, Кисляков Валерий Александрович, Яшина Любовь Петровна

    BACKGROUND: Surgical site infections are common postoperative complications that often develop after hospital discharge. Timely diagnosis and optimal treatment choice are crucial for clinical success and cost-effectiveness in surgical site infection treatment. Computer vision and artificial intelligence methods have demonstrated their effectiveness in analyzing chronic wounds, but their applicability to postoperative wound assessment remains poorly understood. AIM: To compare the diagnostic accuracy of various machine learning models for the classification of surgical wound images to determine the presence or absence of surgical site infections in trauma and orthopedic patients. METHODS: The study included patients aged ≥18 years who were hospitalized at two clinical centers for joint replacement, metal osteosynthesis, spine decompression-stabilization, or other interventions, or for the treatment of surgical site infections following these procedures. The following machine learning algorithms were used for infection recognition: support vector machines (SVMs), logistic regression (LR), random forests (RF), convolutional neural networks (VGG16 + CNN), and a model with an attention mechanism (YOLO 11s-cls). RESULTS: The study group included 183 patients with surgical site infections, and the control group consisted of 115 patients without surgical site infections. A total of 512 surgical wound images (292 with surgical site infection and 220 without infection) obtained from 298 patients were included in the study. After augmentation, the training set comprised approximately 2500 images. The YOLO 11s-cls model demonstrated the best metrics on the test set: sensitivity 91.2%, accuracy 91%; F1-score 89%. However, the differences between this model and the VGG16 + CNN neural network in terms of sensitivity, specificity, and accuracy were not significant. For the remainig models, sensitivity ranged from 69.6% (RF) to 87% (VGG16 + CNN), accuracy ranged from 68% (RF) to 85% (VGG16 + CNN), and F1-score ranged from 66% (RF) to 83% (VGG16 + CNN). CONCLUSION: The potential of using machine learning methods for remote monitoring of surgical wounds in trauma and orthopedic patients was confirmed. The developed models can be used to create a multimodal system for assessing and monitoring wound infection after surgical interventions.

  • Open Access

    Decision support system for the treatment of surgical site infections in trauma and orthopedic patients

    by Anton G. Nazarenko, Elena B. Kleimenova, Natalia I. Sobur, Archil V. Tsiskarashvili, Mikhail A. Dronov, Vitaly A. Otdelenov, Ekaterina S. Yurchenkova, Liubov P. Yashina, Яшина Любовь Петровна

    BACKGROUND: Despite the availability of clinical guidelines for managing patients with surgical site infections, data on the adherence of real clinical practice to these protocols in Russian hospitals remain limited. Data on how noncompliance with clinical guidelines influences outcomes in patients with surgical site infections are also limited. AIM: This study aimed to develop a clinical decision support system for personalized selection of surgical strategies and antibiotic therapy in surgical site infections, and to evaluate adherence of the provided therapy to clinical guidelines for the diagnosis and treatment of surgical site infections in trauma and orthopedic practice. METHODS: It was a single-center retrospective continuous uncontrolled observational study of data from patients hospitalized at the N.N. Priorov National Medical Research Center of Traumatology and Orthopedics, Ministry of Health of Russia, between January 2024 and May 2025. Eligible patients had clinical signs of postoperative wound inflammation, fever (body temperature 37.5 °C) persisting for three or more days after surgery, a positive wound culture, or a consultation with a clinical pharmacologist regarding antibiotic therapy. The study endpoints included compliance of surgical strategy and antibiotic therapy with clinical guidelines (choice of drug, dosage, and duration). Adherence to therapy was assessed by comparing prescriptions with approved surgical site infection treatment protocols using the clinical decision support system. Compliance with guidelines was analyzed in the overall cohort as well as in subgroups of acute and chronic infection. RESULTS: The study included 90 patients, of whom 84 met surgical site infection criteria (31 with acute and 53 with chronic surgical site infections). Full adherence to clinical guidelines was observed in 20 cases (24%) (13 acute, 7 chronic). In 25 cases (30%), the choice of antimicrobial agent did not comply with the recommended one; in 7 cases (8%), the drug was appropriate, but the regimen did not comply with dosing recommendations. In 58 cases (64%), the duration of antibiotic therapy was shorter than recommended, with 16 patients (28%) not receiving antibiotics at the outpatient stage. Patients with chronic infections were significantly more likely to have insufficient antibiotic therapy duration (p 0.0001) and inappropriate drug choice (p = 0.036) compared with those with acute infection. Microbiological cultures were obtained from 82 patients, of whom 33% were already receiving antibiotic therapy at the time of sampling. CONCLUSION: In a trauma and orthopedic hospital setting, there is a clear need for clinical decision support systems based on clinical guidelines, seamlessly integrated into clinicians’ workflow, particularly regarding antibiotic therapy strategies.

  • Open Access

    Retrospective multicenter study of injuries associated with personal mobility devices in Moscow

    by Anton G. Nazarenko, Maxim V. Naumenko, Aleksandr A. Ochkurenko, Elena B. Kleimenova, Liubov P. Yashina, Stanislav E. Maletin, Damir R. Davydov, Aleksandra V. Rogoschenkova, Яшина Любовь Петровна

    BACKGROUND: The use of personal mobility devices is associated with a high risk of injury, with the number of injuries increasing in direct proportion to the growing popularity of personal mobility devices and the number of trips taken. This situation necessitates research into injury patterns to enable the implementation of preventive measures and the development of unified treatment algorithms. Russian scientific sources mainly present isolated data based on small samples, which complicates analysis and systematization. AIM: This study aimed to assess the level, distribution, and temporal changes of injuries associated with personal mobility devices in the largest Russian metropolis (Moscow); to identify trends and risk factors affecting the safety of personal mobility devices operation; and to initiate the systematization of clinical data on personal mobility device-related injuries across the Russian Federation. METHODS: It was a multicenter, non-randomized, retrospective, anonymized study of personal mobility device-related injuries. The analysis was based on reports and medical records from patients treated in 5 outpatient and 7 inpatient facilities of the Moscow City Health Department. The study sample comprised 4678 patients aged 18–89 years, divided into two groups: the first included 3302 patients (70%) who received outpatient medical care or refused hospitalization, and the second included 1376 patients (30%) who were admitted for inpatient treatment. Methods of retrospective analysis, stratification, and statistical data processing were applied. RESULTS: Hospitalized patients accounted for 1376 cases (30%), with a mean age of 35 years (range: 18–86). Most participants were of working age, of both sexes, with no marked gender predominance. However, hospitalized patients were significantly older than outpatients (p 0.05). The gender distribution did not differ between groups (p 0.05). In the group hospitalized for inpatient treatment, injuries resulting from bicycles, e-bikes, scooters, e-scooters and other personal mobility devices (rollerblades, unicycles, Segways, hoverboards), which constituted less than 1% of the overall distribution, accounted for 330 (24%), 69 (5%), 234 (17%), 660 (48%), and 83 cases, respectively. The mean length of hospital stay was 4.76 days. Among outpatients, injuries resulting from bicycles, scooters, e-scooters and other personal mobility devices (rollerblades, unicycles, Segways, hoverboards), which constituted less than 1% of the overall distribution, accounted for 765 (23%), 667 (20%), 1583 (48%), and 287 (9%) cases. Across all personal mobility devices, e-scooters ranked first in injury incidence, dominating the shared mobility market over the past five years. Falls accounted for 79% of all injuries. By the nature of the injuries, in the outpatient group, isolated injuries accounted for 2628 cases (79%), whereas multiple, combined, or associated injuries accounted for 674 (21%) cases. In inpatients, isolated injuries occurred in 1045 (76%) cases, and multiple, combined, or polytrauma injuries occurred in 331 (24%) cases. CONCLUSION: This study analyzed personal mobility device-related injuries treated in Moscow healthcare facilities and determined the distribution of injuries by etiology, nature, and type of medical care provided. The study represents the first step toward systematizing clinical data on personal mobility device-related injuries in Russia for the subsequent development of optimal solutions for prevention and treatment.

  • Open Access

    Prospects for the generative artificial intelligence application in surgery, traumatology and orthopedics

    by Anton G. Nazarenko, Elena B. Kleimenova, Nodari M. Kakabadze, Alexey I. Molodchenkov, Liubov P. Yashina, Яшина Любовь Петровна

    The review considers the use of generative artificial intelligence technologies in surgery, traumatology and orthopedics. Definitions of key generative artificial intelligence technologies are given, as well as the difference between discriminative and generative models of artificial intelligence. An analysis of publication activity on the use of generative artificial intelligence in surgery, traumatology and orthopedics in world macroregions is conducted. The potential role of various generative artificial intelligence models at the preoperative, intraoperative and postoperative stages of healthcare is analyzed. Data on the results of clinical application of generative artificial intelligence and the most common problems associated with the practical use of various generative artificial intelligence applications are provided including issues of quality and safety of surgical care. The review proposes potential solutions and research directions to address these problems.


  • Open Access

    Practical issues in postoperative pain management: pain assessment and personalized choice of analgesics (review)

    by Vitaly A. Otdelenov, Natalia P. Denisenko, Natalia I. Sobur, Elena B. Kleimenova, Anton G. Nazarenko, Israil Magomed-Khadzhievich Gairabekov, Karin B. Mirzaev, Dmitry A. Sychev

    This review focuses on the critical aspects of rational postoperative pain management in surgical hospitals, emphasizing a multimodal, opioid-sparing approach. We discuss the principles of evidence-based analgesia, including the importance of using validated pain scales for consistent pain assessment and defining pain thresholds for initiating and modifying analgesic therapies. Defining target pain levels is crucial for guiding therapeutic interventions and achieving optimal pain relief. This paper describes a systematic and multi-modal approach to analgesic selection, emphasizing the combination of non-opioid analgesics like paracetamol (acetaminophen) and non-steroidal anti-inflammatory drugs (NSAIDs) with opioid analgesics, regional analgesia, and non-pharmacological interventions. Continuous evaluation of pain levels and response to therapy is essential for ensuring effective pain control and minimizing the risk of complications. The review highlights the potential benefits of preoperative analgesia in reducing opioid requirements and the incidence of chronic neuropathic pain. We discuss the unique challenges of pain management in children, emphasizing the need for age-appropriate dosing and administration routes, taking into account the physiological differences between adults and children. The review emphasizes the crucial role of individualizing pain management strategies based on patient characteristics including age, co-morbidities, drug tolerance, and genetic factors. We discuss the impact of genetic polymorphisms in CYP2C9 and CYP2D6 enzymes on NSAID and opioid metabolism in relation to optimizing both efficacy and safety. This review provides a comprehensive overview of the principles and practices for rational postoperative pain management, aiming to improve pain control, reduce opioid use, minimize adverse drug reactions, and enhance patient satisfaction and quality of life following surgical interventions. Implementing standardized pain assessment tools and protocols, utilizing a multi-modal analgesic approach, emphasizing preventive analgesia strategies, considering genetic factors in individual treatment plans, and prioritizing patient education and shared decision-making are key recommendations for healthcare professionals involved in the postoperative management of surgical patients. By adopting these evidence-based recommendations, we can strive to deliver optimal pain relief and enhance the overall recovery experience for patients.


  • Open Access

    Information technologies for supporting prevention, diagnosis and management of surgical site infections in trauma and orthopedic patients

    by Anton G. Nazarenko, Elena B. Kleimenova, Mikhail A. Dronov, Dmitry S. Gorbatyuk, Nodari M. Kakabadze, Archil V. Tsiskarashvili, Natalia P. Gerasimova, Ekaterina S. Yurchenkova, Liubov P. Yashina, Горбатюк Дмитрий Сергеевич, et al.

    BACKGROUND: Surgical site infections (SSIs) are one of the most common preventable healthcare-associated infections, with a significant socioeconomic burden. Information technology, particularly clinical decision support systems, has been shown to improve patient safety. However, there have been few publications on the use of these systems for the prevention and treatment of SSIs. AIM: To develop a multifunctional patient safety information system (PSIS) for the prevention, diagnosis, and treatment of SSIs in traumatic and orthopedic surgery. MATERIALS AND METHODS: A PSIS for traumatic and orthopedic surgeons was developed and implemented in the N.N. Priorov National Medical Research Center of Traumatology and Orthopedics in late 2023 to early 2024. The PSIS includes two user systems: 1) an analytical and authoring system (PSIS-Manager) for subject matter experts who develop scenarios; 2) a medical system for patient data entry and access to guidelines. RESULTS: A PSIS prevention scenario was developed based on the guidelines of the National Association of Infectious and Non-Infectious Disease Control Professionals. It includes 11 recommendations for the preoperative stage, 30 for the perioperative and intraoperative stages, 33 for SSI prevention during surgery, and 7 for the postoperative stage. A surgical antibiotic prophylaxis scenario includes 24 recommendations. Decision-making algorithms for the treatment of SSIs are based on the guidelines of the American Academy of Orthopaedic Surgeons (AAOS). Decision-making involves 6 factors providing 264 various clinical scenarios, with 9 decision options. CONCLUSION: Scenario-based protocols can be used to support decision-making on patient management strategy, as well to control compliance with SSI prevention and treatment guidelines. Further perspectives on PSIS development in terms of SSI diagnosis and treatment include the use of artificial intelligence technologies to aid in the diagnosis of wound infections and the selection of treatment options.


  • Open Access

    Detection of anticoagulant medication errors by triggers

    by V A Otdelenov, E B Kleymenova, M D Nigmatkulova, S A Payushchik, O D Dukhanina, L P Yashina, D A Sychev

    Background. Medication errors can cause preventable adverse events. For example, inappropriate use of anticoagulants (AC) can result in bleeding and thromboembolic complications. Detection and analysis of AC medication errors allow to reveal deficiencies in the safety systems in healthcare organizations. Aim. The study was aimed to develop a method of systematic detection of anticoagulant medication errors for consequent audit, analysis and development of medication safety improvement measures. Materials and methods. The study was conducted in the multidisciplinary hospital and included 4924 patients admitted from January 2019 to December 2021 who received AC. Three laboratory triggers (international normalized ratio ≥4, serum creatinine ≥133 μmol/l, and glomerulofiltration rate 30 ml/min/1.73 m2) helped to reveal 4304 cases. Their matching with patient’s data helped to develop combined triggers. Two clinical pharmacologists reviewed all cases identified by combined triggers for checking medication errors. The trigger was considered positive when anticoagulant medication error was detected in the history selected by combined trigger. Results. Of the 4924 patients 253 (5.3%) were selected by combined triggers. Combined trigger allowed to reduce the amount of medical health records audit by 97.3%. Medication errors were detected in 137 patients. Positive predictive value of selected combined triggers varied from 0 to 63.9%. Aggregated positive predictive value of all combined triggers amounted to 54.2%. AC medication errors were detected in 2.8% patients. Conclusion. Method of systematic detection of AC medication errors using combined triggers in all hospitalized patients receiving AC allowed to reveal typical medication errors for consequent analysis and elaboration of measures to reduce preventable patient harm in healthcare settings.

  • The Impact of Medication Reconciliation on Continuityof Drug Therapy during Perioperative Period

    by Maria D. Nigmatkulova, E. B. Kleymenova, Liubov P. Yashina, Dmitry A. Sychev, Клейменова Елена Борисовна, Яшина Любовь Петровна

    Background. Medication errors and resultant adverse drug events (ADEs) often occur during transitions of care. Up to 67% of in-patients prescriptions have at least one unintentional medication discrepancy with previously prescribed therapy. The proportion of clinically significant medication discrepancies is 1159%. Studies from the developed countries demonstrated the effectiveness of medication reconciliation in reducing medication errors, ADEs and healthcare resource utilization. There is a necessity to conduct medication reconciliation studies within Russian current clinical practice to develop effective medical care quality and patients safety programs. Aims to evaluate the impact of pharmacologist-led medication reconciliation on the frequency and structure of unintentional medication discrepancies and potential ADEs at hospital admission and discharge. Methods. Standard care was compared to medication reconciliation led by a clinical pharmacologist in a prospective randomized trial of 410 elective surgical patients. Medication discrepancies at hospital admission and discharge were identified and reconciled. Clinical outcomes were evaluated by reviewing electronic health records. Results. In the intervention group the frequency of unintentional discrepancies at hospital admission decreased from 32.68 to 16.86%, the proportion of patients with at least one unintentional discrepancy decreased from 64.9 to 44.9%, the number of discrepancies per patient decreased from 1.5 to 0.66. The incidence of discrepancies at hospital discharge decreased from 82.90 to 43.29%, the proportion of patients with discrepancies decreased from 95.61 to 52.68%, the average number of discrepancies per patient decreased from 2.79 to 1.67. Medication reconciliation led by clinical pharmacologist decreased the frequency of unscheduled out-patient visits after discharge from 7.32 to 2.93%. The determined risk factors for unintentional discrepancies at hospital admission were: prescribing of cardiovascular, endocrine drugs and those affecting the central nervous system. Both at admission and discharge medication reconciliation was the significant factor reducing the risk of unintentional discrepancies. Conclusions. Medication reconciliation at hospital admission and discharge reduces the frequency of unintentional discrepancies in drug prescriptions by 16 and 40%, respectively. The implementation of medication reconciliation into clinical practice reduces unscheduled out-patient visits after hospital discharge.

  • User-Configurable Structure of Decision Support System for Clinical Risk Management

    by Elena Kleymenova, Mikhail Dronov, Liubov Yashina, Svetlana Payushchik, Maria Nigmatkulova, Vitaly Otdelenov