Развитие диагностических технологий в медицине, в частности в офтальмологии, позволило изучить человеческий глаз не только на гистологическом, но и на клеточном уровне. Роль нервной системы в работе зрительного анализатора связана не только с обеспечением двигательной, защитной и сенсорной функций, а прежде всего, с обеспечением самого зрительного процесса. Нейродегенеративные заболевания занимают особое место среди системных патологий. В настоящее время развитие данного процесса связывают с нейровоспалением, роль которого подтверждена и при глаукоме. По этой причине достижение целевого уровня внутриглазного давления не всегда гарантирует стабилизацию дегенеративного процесса. Ввиду этого препараты-нейропротекторы рекомендуется применять у любого пациента с глаукомой с учетом особенностей патогенеза заболевания. Благодаря своим антиоксидантным и нейропротекторным свойствам этилметилгидроксипиридина сукцинат (Мексидол) широко используется в офтальмологической практике, в частности при лечении пациентов с глаукомной оптической нейропатией и заболеваниями сетчатки. Результаты исследований продемонстрировали эффективность Мексидола в отношении замедления нейродегенерации и стабилизации зрительных функций у больных первичной открытоугольной глаукомой благодаря антигипоксантным, антиоксидантным и мембраностабилизирующим свойствам препарата и его положительному влиянию на нейромедиаторный баланс и глазной кровоток.

Background: Ovarian cancer is characterized by high mortality rates, primarily due to diagnosis at late stages. Current biomarkers, such as CA125, have demonstrated limited efficacy for early detection. While high-dimensional proteomics offers a more comprehensive view of systemic biology, the analysis of such data, where the number of features far exceeds the number of samples, presents a significant computational challenge. Methods: This study utilized a nested case–control cohort of longitudinal pre-diagnostic serum samples from the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) profiled for eight candidate ovarian cancer biomarkers (CA125, HE4, PEBP4, CHI3L1, FSTL1, AGR2, SLPI, DNAH17) and 92 additional cancer-associated proteins from the Olink Oncology II panel. We employed a Synolitic Graph Neural Network framework that transforms high-dimensional multi-protein data into sample-specific, interconnected graphs using a synolitic network approach. These graphs, which encode the relational patterns between proteins, were then used to train Graph Neural Network (GNN) models for classification. Performance of the network approach was evaluated together with conventional machine learning approaches via 5-fold cross-validation on samples collected within one year of diagnosis and a separate holdout set of samples collected one to two years prior to diagnosis. Results: In samples collected within one year of ovarian cancer diagnosis, conventional machine learning models—including XGBoost, random forests, and logistic regression—achieved the highest discriminative performance, with XGBoost reaching an ROC-AUC of 92%. Graph Convolutional Networks (GCNs) achieved moderate performance in this interval (ROC-AUC ~71%), with balanced sensitivity and specificity comparable to mid-performing conventional models. In the 1–2 year early-detection window, conventional model performance declined sharply (XGBoost ROC-AUC 46%), whereas the GCN maintained robust discriminative ability (ROC-AUC ~74%) with relatively balanced sensitivity and specificity. These findings indicate that while conventional approaches excel at detecting late pre-diagnostic signals, GNNs are more stable and effective at capturing subtle early molecular changes. Conclusions: The synolitic GNN framework demonstrates robust performance in early pre-diagnostic detection of ovarian cancer, maintaining accuracy where conventional methods decline. These results highlight the potential of network-informed machine learning to identify subtle proteomic patterns and pathway-level dysregulation prior to clinical diagnosis. This proof-of-concept study supports further development of GNN approaches for early ovarian cancer detection and warrants validation in larger, independent cohorts.

Background: Post-COVID-19 condition (PCC) affects many survivors, with evidence of sex-specific differences in prevalence and symptom profiles. However, few prediction studies have examined whether sex-stratified models improve prediction or generalize across sexes. This study aimed primarily to develop and compare sex-stratified machine learning models for PCC prediction using routinely available baseline variables, and secondarily to assess cross-sex generalizability and adversarial robustness. Methods: We analyzed a prospective longitudinal cohort of 1006 adults hospitalized with COVID-19 at Sechenov University Hospital Network (Moscow, Russia). Demographics, smoking status, and pre-existing comorbidities were extracted from medical records, and PCC status was assessed at 6-month follow-up. Machine learning models—including classical algorithms and graph-based neural networks—were trained separately for males and females. Cross-sex validation evaluated generalizability, variable importance aided interpretation, and adversarial perturbations assessed model robustness. Results: PCC prevalence was higher in females (53.9%) than males (39.1%). Overall predictive performance was modest across all models, with AUC values ranging approximately 0.50–0.61. Graph-based models achieved the highest discrimination, with the best AUC reaching approximately 0.61, while classical approaches provided limited predictive value. Cross-sex validation showed minor asymmetry: models trained on male data performed slightly better on female cases than vice versa. Adversarial testing revealed sensitivity of all models to input perturbations. Conclusions: Demographics and comorbidities alone provide insufficient information for reliable PCC prediction. Modest sex-specific differences in model generalizability suggest distinct, sex-associated PCC phenotypes, but richer multimodal data—including clinical biomarkers, wearable-derived measures, and patient-reported outcomes—will be required to develop clinically useful and equitable predictive models. Sex-stratified approaches should be considered in future post-viral syndrome prediction studies.

Background/Objectives: Universal surgical risk calculators are not validated for use and poorly predict postoperative morbidity and mortality for women undergoing gynaecological oncology surgery. This adversely affects communication of risk resulting in poorly informed decisions and missed opportunities for medical optimization to mitigate risk preoperatively. We present the development and external validation of our novel GO SOAR surgical risk calculator for use to preoperatively predict postoperative thirty-day surgical morbidity and mortality in relation to gynaecological oncology surgeries. Methods: New logistic regression models were developed using the GO SOAR1 training cohort (n = 1811) and externally validated in an independent prospective cohort (n = 416) for two outcomes: thirty-day postoperative mortality (alive versus dead) and thirty-day postoperative morbidity (any complication (Clavien–Dindo I–V) versus none). Performance of the GO SOAR models was compared against established all-purpose surgical risk calculators (SORT/POSSUM/P-POSSUM/NSQIP). Model discrimination was assessed with sensitivity calculated at a clinically significant prespecified specificity threshold of 90%. Results: For mortality, AUROC was 0.752 (95% CI 0.570–0.935) for GO SOAR full and 0.795 (95% CI 0.660–0.930) for GO SOAR condensed; corresponding sensitivities at 90% specificity were 57.1% and 42.9%. For morbidity, AUROC was 0.698 (95% CI 0.641–0.755) and 0.703 (95% CI 0.646–0.760) for the full and condensed models, respectively, with sensitivities of 30.4% and 32.1% at 90% specificity. Conclusions: The GO SOAR model using a data-driven, gynaecological-oncology-specific approach at 90% specificity, achieved the highest observed sensitivity among the evaluated calculators. Accurate surgical risk predictions are crucial for major oncological surgery, where complications can diminish quality of life and affect long-term cancer survival. A model such as the GO SOAR surgical risk calculator that uses readily available preoperative data, regardless of income setting, is essential in reducing global disparities in surgical outcomes.

To understand the current situation with the genetic epidemiology of primary congenital glaucoma (PCG) in the world, as well as to identify genotype-phenotype correlations, 87 publications were analyzed from the Pubmed and elibrary databases, mainly the latest 5 years. This publication presents the first part of the analysis, covering Arab and African countries, Turkey, the USA, Brazil, Australia and New Zealand. The second part of the review, which will be published in early 2024, will focus on data from East and South Asia, Western Europe and Russia. The highest incidence of PCG is typical for Arab countries, which is associated with the presence of a significant number of consanguineous marriages in these countries. In most countries, boys predominated among patients with PCG, and the process was often bilateral. In Australia female patients predominate. The PCG phenotype is more severe in patients with CYP1B1 mutations (early onset, bilateral involvement) than in patients without mutations, and in patients with consanguineous parents than in children from unrelated marriages. Among the genetic causes of PCG in different populations, the most common (from 30 to 86 % of cases) in Arab countries, Brazil and Turkey are mutations in the CYP1B1 gene. The spectrum of mutations varies depending on the region. In Australia and New Zealand, biallelic mutations were identified in the CYP1B1, CPAMD8 and COL18A1 genes, as well as heterozygous mutations in the TEK, FOXC1 and ANGPT1. Mutations in rare genes were also found in patients with PCG from Lebanon (ANGPT1 (p.K186N), Mauritania (NTF4 (c.601T>G, p.Cys201Gly) and WDR36 (c.2078A>G, p.Asn693Ser)). In the USA, the influence of the thrombospondin gene (THBS1) on the development of PCG is being studied. Based on the results of genetic and epidemiological studies, specific treatment, rehabilitation and preventive programs can be developed, which will determine the success of PCG therapy in children, preserving their visual functions.

This publication continues a review published in December 2023 about the genetic epidemiology and genotype-phenotype correlations in primary congenital glaucoma (PCG) worldwide, and presents data from East and South Asia, Western Europe and Russia. Analysis of the incidence of PCG data in these regions shows that the rates for Western Europe, the European part of Russia and East Asia are comparable and average 1:10,000–18,000 newborns and higher in ethnically mixed groups. On the territory of South Asia, the minimum number of children with PCG was recorded in Nepal, and the maximum in India. The majority of researchers talk about the predominance of bilateral process in 62 to 99 % cases. As a rule, there is a slight predominance of boys among patients with PCG; authors from India and France report approximately the same number of boys and girls, and German scientists have found a higher prevalence of PCG among girls. If we talk about the phenotype of the disease, almost all researchers note a more severe course in subgroups with the presence ofCYP1B1mutations than in patients who do not have a mutation in this gene. As for the genetic causes of PCG, the most common are mutations in theCYP1B1gene, accounting for up to a third of cases in India, Pakistan, Western Europe and the European part of Russia. In East Asian patients, mutations in theCYP1B1gene are most likely not the main ones in the pathogenesis of PCG, because are found much less frequently and often only in one allele. Chinese scientists say that it is necessary to pay attention to mutationsTEKandZC2HC1C,VPS13D,PGF.