Relevance. The desire to improve and optimize the results of surgical treatment of coronary heart disease (CHD), along with the observed integration of artificial intelligence methods into healthcare, creates prerequisites for exploring the possibilities of machine learning meth Relevance. The desire to improve and optimize the results of surgical treatment of coronary heart disease (CHD), along with the observed integration of artificial intelligence methods into healthcare, creates prerequisites for exploring the possibilities of machine learning methods for predicting adverse outcomes after cardiac surgery. The purpose of our study was to evaluate and compare the accuracy of predicting death after CABG surgery using machine learning methods and the recommended cardiac risk assessment scale EuroSCORE 2. Materials and methods. Based on the analysis of depersonalized medical data on the outcomes of coronary artery bypass surgery in 2,826 patients with coronary artery disease (survivors — 2,785, deceased — 41), using machine learning methods (logistic regression (LR), LightGBM, XGBoost, CatBoost, boosting model), prognostic models were developed that assess the risk of intrahospital death after intervention. The forecasting efficiency of the obtained models was compared with the forecasting results of the EuroSCORE 2 scale. To evaluate the performance of the models, the metrics recommended for the analysis of unbalanced data were used: precision, recall, specificity, F1-measure, ROC-AUC. Results. The model developed with the help of LR had the maximum recall (0.88), but at the same time significantly overestimated the risk of death (precision — 0.03). F1-measure for the LR model was 0.06, ROC AUC — 0.77. Gradient boosting models (LightGBM, XGBoost, CatBoost), in comparison with LR, had higher indicators of precision, recall, specificity, F1-measures and AUC. At the same time, the best quality metrics were observed in the boosting model (BM), which combined LR and gradient boosting models. BM performance indicators: precision — 0.67, recall – 0,50, F1-measure — 0.57, specificity — 1.0, ROC-AUC — 0.85. The EuroSCORE 2 risk model showed extremely low efficiency in predicting death in the study sample: precision — 0.143, recall — 0.125, F1-measure — 0.133, specificity — 0.97, ROC-AUC — 0.47. Conclusion. Machine learning (ML) methods are promising in predictive analytics in cardiac surgery. In our study, predictive models based on ML showed an advantage in the accuracy of calculating the risk of hospital death after CABG in comparison with the classic EuroSCORE 2 model. To obtain an optimal risk model adapted to the conditions of application in the Russian Federation, largescale multicenter studies are needed.

Noncommunicable diseases (NCDs) are the leading cause of mortality worldwide, with a vast majority of them having a multifactorial etiology, including genetics. That is why implementation of polygenic risk scores (PRS) in clinical practice with the aim of development of individual strategies for NCDs prevention seems promising. In this review we will discuss development and possible implications of PRSs.

The article deals with an overview of modern concepts of personalized neurology — part of a continuum of personalized medicine. A brief historical and terminological essay is given, as well as a rationale for the importance of such an approach. Examples of promising areas of research on a wide range of biomarkers are described: biochemical, neurophysiological, angioneuroimaging, pathomorphological, etc. The issues of integrating the results of fundamental work in the field of neuroscience into clinical practice for the individualization of treatment are discussed. A general review of existing approaches to personalization in the diagnosis and therapy of the most significant neurological diseases is given: cerebrovascular pathology, epilepsy, demyelinating and neurodegenerative diseases. Future implications for the development of this area are identified and discussed, examples are described that reflect certain socio-economic aspects of personalization in neurology. The conclusion also lists a number of challenges facing this new direction in healthcare.

Modern medical science increasingly emphasizes a personalized approach to the treatment of various diseases. At the same time, however, population-based research remains an integral part of medical practice. Do these two approaches compete or can they be synergistic? In the presentation of Natalia G. Mokrysheva, Corresponding Member of the Russian Academy of Sciences, Director of Endocrinology Research Center on May 19, 2023, within the framework of the VI Innovative St. Petersburg Medical Forum, the history of the population approach, the opportunities and challenges of personalization were discussed, and a brief overview of the interaction of these two approaches in modern Russian endocrinology was given.

The article presents the concept of development of research in the field of pharmacogenetics and personalized medicine in the Russian Federation and ways of introducing “omics” technologies into widespread clinical practice.

The review analyzes the literature data on the role of circulating microRNAs in human kidney cancer. Basic information about the mechanisms of biogenesis of microRNAs is given, their regulatory role is considered. Particular attention is paid to the prospect of using miRNAs as diagnostic and prognostic biomarkers of kidney cancer, taking into account the conflicting data found in various studies.

Background. Among malignant neoplasms in HIV-infected patients lymphomas occupy a special place due to the high incidence, course characteristics, and difficulties that arise during diagnosis and during antitumor drug therapy. Hodgkin’s lymphoma (HL) is not an AIDS indicating disease, but the risk of its development in people infected with HIV is 5–25 times higher than the incidence of HL in the general population. Prior to the use of antiretroviral therapy, the results of standard chemotherapy in HIV-infected patients with HL were significantly worse than in HIV-negative patients. One of the main requirements for drug treatment of this group of patients is the simultaneous use of antiretroviral therapy and chemotherapy. The aim was to study the clinical characteristics and results of treatment of HL in the presence of HIV infection. Materials and methods. The analysis included 24 HL patients with HIV infection who received treatment in the Department of Radiation and Drug Therapy of Hemoblastoses of the MRRC in the period from 2018 to 2022. Treatment program selection was in accordance with the HL treatment protocol developed at our Center. Patients received 4–6 cycles of ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) or 6 cycles of BEACORP (bleomycin, vepesid, doxorubicin, cyclophosphamide, vincristine, dacarbazine, prednisolone) chemotherapy according to the stage of the disease and the risk factors generally accepted for patients with HL. The response to therapy was assessed according to the Lugano-2014 criteria. Descriptive statistics methods were used. Overall survival and progression-free survival were analyzed using the Kaplan-Meier method. Results. HL occurring against the background of HIV is most often represented by a widespread nodal and extranodal lesion, accompanied by symptoms of intoxication (B-symptoms). The use of standard CT regimens as induction therapy for HL in the presence of HIV makes it possible to obtain satisfactory immediate and long-term results of treatment. In our study complete and partial responses were achieved in 94.1 %. With a median follow-up of 12 months survival without progression and overall survival were 75 % and 100 % respectively.

Purpose. Description of the results of the phase I clinical study of albumin microspheres 20– 40 microns, labeled ¹⁸⁸Re (Hepatoren). Materials and methods. From June 2022 to August 2023, 20 patients with inoperable liver tumors underwent radioembolization of the liver with ¹⁸⁸Re albumin microspheres 20–40 μm at the A. F. Tsyba MRSC — a branch of the Federal State Budgetary Institution “NMIC of Radiology” of the Ministry of Health of the Russian Federation. Results. The technical success of the operation is 100 %. After the introduction of ^99mTc-MAA and the implementation of the SPECT/CT scan of none of the patients revealed the accumulation of radiopharmaceuticals in the organs of the gastrointestinal tract; bypass surgery to the lungs ranged from 1 % to 4.5 %. Grade I hematological toxicity was observed in 15 (75 %) patients, grade II — in 3 (15 %) patients, grade III — in 2 (10 %) patients. All the studied patients had grade II hepatological toxicity. Good tolerability was observed in 10 (50 %) patients, satisfactory tolerability — in 7 (35 %) patients, unsatisfactory tolerability — in 3 (15 %) patients. There were no pronounced complications during the follow-up of patients of the first phase of the study. Conclusion. Our first experience of using domestic microspheres has demonstrated acceptable indicators of toxicity and tolerability, which allows us to continue working in this direction.

In this study, we described a case that illustrates the clinical significance of the hook-effect, a laboratory phenomenon when the measured hormone level is falsely lowered due to the binding of signaling antibodies with high analyte concentration. In a patient with a newly diagnosed 2.7 x 1.7 x 2.1 cm pituitary macroadenoma, laboratory testing revealed an increase in prolactin level to 878.6 mMU/ml (105.0–540.0 mMU/ ml). The abnormality was interpreted as a secondary increase hyperprolactinemia due to pituitary stalk compression, and the patient was referred for surgery. Upon admission to the hospital, the prolactin level was measured repeatedly using a more modern test system and the level was 2 296.0 ng/ml (4.0–15.2) without detecting a significant amount of macroprolactin. The degree of hyperprolactinemia indicated the presence of macroprolactinoma. The patient was discharged from the neurosurgical department with a recommendation to start dopamine agonists treatment, which showed an excellent biochemical response. A personalized approach to the differential diagnosis of hyperprolactinemia, which takes into account not only a wide range of pathological conditions, but also a number of laboratory phenomena, reduces the risks of unjustified surgical interventions in patients with prolactinomas.