Search Results (475)

Background: Traumatic brain injury (TBI) is a leading cause of disability but one of the least recognized health problems in the world, affecting up to 69 million people annually. The associated lifelong disability in survivors, the loss of economic productivity, and being a risk factor for dementia consume 0.5% of global economic activity. Yet TBI is still largely invisible in national surveillance systems and not well represented in chronic disease frameworks. Consequently, governments are not equipped to provide proportional financing of acute care and long-term care of survivors, nor to build health care systems and resources for improving outcomes of TBI through policy frameworks targeting prevention, treatment, and equitable access. Objective: This commentary aims to provide a comprehensive picture of the global effort to formally recognize TBI as a notifiable and chronic condition, including the justifications for recognition, the formation of an international coalition of stakeholders, and the strategic plan for resolution at WHA79 of the World Health Assembly, one of the first concerted multinational efforts that occurred as a side event during the 78th World Health Assembly (WHA78) in May 2025. Methods: This commentary integrates information from epidemiological studies, global registries, and testimonies from people with lived experience of TBI. We analyze these data to develop policy needs and corresponding initiatives to address key needs. These include coordinated efforts to advocate change, such as technical briefings, consultations with stakeholders, and storytelling led by survivors, all of which informed and formed a part of the WHA78 side event. Our efforts have garnered wide, multi-sector support. Results: The WHA78 side event showed that ministries of health, neurosurgical, neurological, and rehabilitation societies, academic researchers, WHO representatives, and survivors all unprecedentedly support the recognition of the importance of TBI, facilitating national policies for its prevention and treatment via standardized surveillance. More than 30 non-governmental groups officially supported the campaign. A sponsoring member state made a public commitment to co-sponsor a WHA resolution, which set the stage for ongoing diplomatic progress and engagement across regions. Conclusion: To improve global brain health equity, access to long-term care, and the resilience of health systems, it is important to recognize TBI as a notifiable and chronic condition. A dedicated WHA resolution would make TBI a part of global health governance, making sure that it is counted, tracked, and dealt with as quickly and comprehensively as possible. It is both a technical necessity and a moral duty to help survivors and families and fight for justice in global health systems. Full article

Background/Objectives: Major Depressive Disorder (MDD) affects over 280 million people worldwide and is a leading cause of disability. Emotional blunting—characterized by a numbing or flattening of emotions—is a significant yet often underrecognized symptom that impairs daily functioning and interpersonal relationships in patients with MDD. It remains unclear whether emotional blunting results primarily from the disorder itself or from antidepressant treatments, especially selective serotonin reuptake inhibitors (SSRIs) and serotonin–norepinephrine reuptake inhibitors (SNRIs). Vortioxetine, a multimodal antidepressant approved for MDD, may help alleviate emotional blunting by modulating neurotransmitters differently than SSRIs. This study investigates the severity of emotional blunting among Hong Kong MDD patients and explores the changes in this symptom with the use of vortioxetine, while also considering anhedonia as a related dimension of reward processing. Methods: This naturalistic, longitudinal observational study in Hong Kong enrolled adults (aged 18 and above) clinically diagnosed with MDD who were initiating vortioxetine treatment for emotional blunting. Patient inclusion was based on independent prescribing decisions by psychiatrists, with informed consent obtained. Data collection comprised one intake interview and the administration of four self-report questionnaires—ODQ, PHQ-9, PDQ-D, SDS, MFI, and SHAPS—at baseline, week 1, week 4, and week 8. Demographic and clinical history data were also recorded. Questionnaires were completed online or via phone, over a study duration of approximately two months. Results: The prevalence of emotional blunting, estimated by the proportion of patients with an ODQ score at or above the clinical cut-off (≥50), was 91.9% at baseline, decreasing to 85.5% at week 1, 77.7% at week 4, and 73.3% at week 8. Significant improvements were also observed in depressive symptoms, cognitive dysfunction, functional impairment, pleasure experience, and fatigue. Conclusions: In this naturalistic observational cohort of patients with MDD who were prescribed vortioxetine, self-reported emotional blunting, depressive symptoms, cognitive dysfunction, functional impairment, and fatigue decreased over eight weeks. Anhedonia scores (SHAPS) decreased to non-significant levels, and clinician-rated Clinical Global Impression scores confirmed a significant reduction in illness severity. Full article

Long-chain non-coding RNAs (lncRNAs) play important regulatory roles in the growth and development of skeletal muscle, but systematic identification and functional studies of lncRNAs related to porcine skeletal muscle development remain limited. Based on a previously constructed panoramic map of porcine skeletal muscle lncRNAs, lncRNA-ssc.37456 was identified as differentially expressed in porcine skeletal muscle before and after birth. Its function and potential mechanisms were investigated using a porcine skeletal muscle regeneration model, a primary skeletal muscle cell differentiation model, and knockdown and overexpression experiments in vitro. lncRNA-ssc.37456 was upregulated on day 7 of regeneration, with expression positively correlated with the muscle differentiation marker MYHC and negatively correlated with the proliferation marker PAX7. During differentiation of porcine primary myoblasts, expression continuously increased, peaking on day 4. Knockdown of lncRNA-ssc.37456 by small interfering RNA (siRNA) significantly increased cell proliferation, upregulated mRNA and protein levels of proliferation-related genes KI67 and PCNA, and increased the proportion of EdU-positive cells. Conversely, expression of differentiation-related genes MYOG and MYHC decreased, and immunofluorescence analysis revealed reduced myotube formation and differentiation index. Overexpression of lncRNA-ssc.37456 promoted differentiation and inhibited proliferation, showing effects opposite to those observed in knockdown experiments. Nucleocytoplasmic fractionation indicated predominant cytoplasmic localization, suggesting potential function through a ceRNA mechanism. An interaction network with miRNAs was constructed based on the miRDB database, indicating a potential miRNA “sponge” regulatory mechanism. These results indicate that lncRNA-ssc.37456 participates in porcine skeletal muscle development by regulating the transition of muscle cells from proliferation to differentiation, providing molecular insights and potential targets for muscle biology research and the molecular breeding of growth traits. Full article

Accurate electricity carbon emission factors are crucial for assessing overall social carbon emissions and achieving China’s “dual carbon” goals. This paper proposes a dynamic correction model that integrates lifecycle extension, power exchange networks, and multi-time-scale decomposition to address the limitations of static carbon emission factors. The model considers factors such as power generation structure, cross-regional transmission, clean energy proportion, line losses, and non-CO₂ greenhouse gas emissions, and achieves dynamic correction at quarterly and monthly scales, enhancing timeliness and regional adaptability. Results show that transmission losses, energy structure, and inter-provincial electricity exchange significantly impact carbon emission factors. For instance, in 2022, line losses in Xinjiang and Gansu raised the electricity carbon emission factor by over 0.06 kgCO₂/kWh. Monthly factors exhibit significant seasonal fluctuations, with some regions showing variations of up to 105% of the annual average. Areas rich in hydropower, such as Yunnan, Sichuan, and Qinghai, experience pronounced fluctuations, highly sensitive to changes in water volume, offering more accurate reflections of carbon emission changes during electricity consumption. This study presents a refined dynamic correction method for electricity carbon emission accounting, providing theoretical support for carbon emission policy development and performance evaluation. Full article

Background: The current rise in multidrug-resistant pathogens highlights the urgent need for the discovery of novel antibacterial agents with potential clinical applications. A considerable proportion of these developed resistances may be attributable to the intrinsic response of bacteria to antibiotic-induced stress conditions in the environment. Consequently, the identification and characterization of genetic alterations in physiological processes in response to antibiotics represent promising strategies for the discovery and characterization of naturally produced novel antibacterial agents. This study investigated the antimicrobial activity of an antimicrobial active isolate Actinomadura coerulea derived from a meerkat fecal sample. Methods: The production of secondary metabolites that potentially compromise bacterial cell wall integrity was confirmed by the induction of promoter activity in whole-cell biosensors in which an antibiotic-inducible promoter was fused to the luciferase cassette. During plate-based biosensor assays, we identified naturally resistant Bacillus subtilis colonies growing in the zone of inhibition around A. coerulea colonies. After these successive rounds of selection, highly resistant spontaneous B. subtilis mutants had evolved that were subjected to whole-genome sequencing. Results: Non-silent mutations were identified in pssA, which encodes a phosphatidylserine synthase; mdtR, as a gene for the repressor of multidrug resistance proteins, and yhbD, whose function is still unknown. A new cinnamycin-like molecule, coerumycin, was discovered based on the physiological role of PssA and comprehensive genomic analysis of A. coerulea. Additional experiments with cell extracts containing coerumycin as well as the cinnamycin-like compound duramycin confirmed that the interaction between coerumycin and the bacterial cell envelope is inhibited by a loss-of-function mutation in pssA. Conclusion: Our approach demonstrates that combining the exploration of niche habitats for actinomycetes with whole-cell biosensor screening and characterization of natural resistance development provides a promising strategy for identifying novel antibiotics. Full article

Background: A childhood burn presents new and unfamiliar challenges to patients and their parents during recovery. These injuries can negatively impact activities such as independence in self-care, participation in physical activity, and social interaction. As such, pediatric burn patients are at risk of poorer quality of life (QoL) outcomes after their burn. In this longitudinal, observational cohort study, we examined the social, demographic, and clinical factors that were associated with a poor QoL at 12 months postburn for pediatric patients aged > 2 years with non-severe burns in Western Australia. Methods: Inpatients were recruited from the pediatric burn unit at Perth Children’s Hospital in Western Australia between February 2021 and September 2022. Demographic and family information (age, sex, postcode, parental education, languages spoken at home) and clinical data (burn cause, TBSA%, location, surgical interventions, length of stay) were collected at baseline. At 6 and 12 months, caregivers completed the Brisbane Burn Scar Impact Profile (BBSIP). Results: A total of 37 caregivers completed the Brisbane Burn Scar Impact Profile (BBSIP). For the child’s QoL, 57% of caregivers reported that some impact remained for overall QoL, 32% for sensory intensity, 46% for sensitivity, 22% for daily living (22%), and 19% for emotional reactions. Parent worry was impacted in 46% of caregivers. Being female was associated with greater long-term impacts, particularly in overall functioning and parental worry. The burn location also influenced outcomes, with injuries to the upper limbs linked to higher sensory intensity and emotional impact. Children from culturally and linguistically diverse (CaLD) backgrounds, indicated by those speaking a language other than English at home (LOTE), demonstrated significantly greater effects across several domains, including overall impact, daily living, appearance, and parent worry. Conclusions: A substantial proportion of children continued to experience impacts from non-severe burns across multiple domains, indicating that even small-area burns can have lasting effects. The factors associated with worse scores were the child being female, the families being linguistically diverse, and upper body burns. Full article

With the increasing requirements for “Green Mine” construction, Cemented Tailings Backfill (CTB) has emerged as the preferred strategy for solid waste management and ground pressure control in underground metal mines. However, full tailings, characterized by wide particle size distribution and high fine-grained content, exhibit complex physicochemical properties that lead to significant non-linear behavior in slurry rheology and strength evolution, posing challenges for accurate prediction using traditional empirical formulas. Addressing the issues of significant strength fluctuations and difficulties in mix proportion optimization in a specific lead–zinc mine, this study systematically conducted physicochemical characterizations, slurry sedimentation and transport performance evaluations, and mechanical strength tests. Through multi-factor coupling experiments, the synergistic effects of cement type, cement-to-tailings (c/t) ratio, slurry concentration, and curing age on backfill performance were elucidated. Quantitative results indicate that solids mass concentration is the critical factor determining transportability. Concentrations exceeding 68% effectively mitigate segregation and stratification during the filling process while maintaining optimal fluidity. Regarding mechanical properties, the c/t ratio and concentration show a significant positive correlation with Uniaxial Compressive Strength (UCS). For instance, with a 74% concentration and 1:4 c/t ratio, the 3-day strength increased by 1.4 times compared to the 68% concentration, with this increment expanding to 2.0 times by 28 days. Furthermore, a comparative analysis of four cement types revealed that 42.5# cement offers superior techno-economic indicators in terms of reducing binder consumption and enhancing early-age strength. This research not only establishes an optimized mix proportion scheme tailored to the operational requirements of the lead–zinc mine but also provides a quantitative scientific basis and theoretical framework for the material design and safe production of CTB systems incorporating high fine-grained full tailings. Full article

Gesture recognition is important for rehabilitation assistance and intelligent prosthetic control. However, surface electromyography (sEMG) signals exhibit strong non-stationarity, and conventional deep-learning models require long training time and high computational cost, limiting their use on resource-constrained devices. This study proposes a Frozen Multi-Convolution Dual-Branch Network (FMC-DBNet) to address these challenges. The model employs randomly initialized and fixed convolutional kernels for training-free multi-scale feature extraction, substantially reducing computational overhead. A dual-branch architecture is adopted to capture complementary temporal and physiological patterns from raw sEMG signals and intrinsic mode functions (IMFs) obtained through variational mode decomposition (VMD). In addition, positive-proportion (PPV) and global-average-pooling (GAP) statistics enhance lightweight multi-resolution representation. Experiments on the Ninapro DB1 dataset show that FMC-DBNet achieves an average accuracy of 96.4% ± 1.9% across 27 subjects and reduces training time by approximately 90% compared with a conventional trainable CNN baseline. These results demonstrate that frozen random-convolution structures provide an efficient and robust alternative to fully trained deep networks, offering a promising solution for low-power and computationally efficient sEMG gesture recognition. Full article

The study aims to develop a method for obtaining a high-performance catalyst for the synthesis of liquid hydrocarbons using the Fischer–Tropsch method based on ultradisperse cobalt powders obtained by the electric explosion method. To determine the catalytic activity of the obtained catalyst samples, the main process parameters, like temperature in the catalyst bed, the process pressure, the feedstock space velocity, and the ratio of reagents in the synthesis gas, were varied. It has been established that highly dispersed cobalt powder obtained by the electrical explosion method is a fairly active catalyst for the synthesis of liquid hydrocarbons via the Fischer–Tropsch process. It has been established that the overall CO conversion rate in the temperature range from 230 to 330 °C ranges from 25 to 90%. However, the formation of the main byproduct of the synthesis, carbon dioxide, is not observed below 270 °C. It was determined that for the developed catalyst sample, the optimal temperature range is from 230 to 260 °C, in which the yield of by-products of synthesis and gaseous hydrocarbons is quite low—the selectivity for methane does not exceed 20%, with the proportion of C₅₊ hydrocarbons in the liquid phase at the level of 80%. The CO conversion rate increases proportionally with growing pressure. It has been established that cobalt nanopowder exhibits high catalytic activity in reactions of liquid hydrocarbon formation with low hydrogen content in the initial synthesis gas. This fact allows us to conclude that it has potential for use in processing gases obtained during the pyrolysis of biomass or other non-traditional sources of synthesis gas, characterized by an H₂:CO ratio of 1:1 to 1.25:1. Catalysts obtained from ultradisperse cobalt powders were shown to be resistant to rapid deactivation under synthesis conditions at operating temperatures for 30 h. During long-term testing, CO conversion remained at 23.5% at 230 °C for the entire duration of the experiment. Full article

Biological decision-making under uncertainty relies on somatic markers, which are affective signals that bias choices without exhaustive computation. This study biomimetically translates the Somatic Marker Hypothesis (SMH) into synthetic somatic markers (SSMs), a minimal and interpretable evaluative mechanism that assigns a scalar valence to compressed environmental states in the high-dimensional discrete grid-world Pixelverse, without modelling subjective feelings. SSMs are implemented as a lightweight Python routine in which agents accumulate valence from experience and use a simple threshold rule (θ = −0.5) to decide whether to keep the current trajectory or reset the environment. In repeated simulations, agents perform few resets on average and spend a higher proportion of time in stable “good” configurations, indicating that non-trivial adaptive behaviour can emerge from a single evaluative dimension rather than explicit planning in this small stochastic grid-world. The main conclusion is that, in this minimalist 3 × 3 Pixelverse testbed, SMH-inspired SSMs provide an economical and transparent heuristic that can bias decision-making despite combinatorial state growth. Within this toy setting, they offer a conceptually grounded alternative and potential complement to more complex affective and optimisation model. However, their applicability to richer environments remains an open question for future research. The ethical implications of deploying such bio-inspired evaluative systems, including transparency, bias mitigation, and human oversight, are briefly outlined. Full article

Platycladus orientalis exhibits significant potential as an antioxidant, anti-inflammatory, and hair growth-promoting ingredient, while the low bioavailability of raw Platycladus orientalis leaves extracts limits their further application. In this study, yeast fermentation was employed to prepare Platycladus orientalis Leaves Yeast Fermented Solution (PYFS). Its performance on human dermal papilla cells (HDPCs) was systematically investigated. The optimal fermentation strain was screened using the methyl thiazolyl tetrazolium (MTT) assay, and Saccharomycopsis fibuligera CICC33226 (SF) was identified as the most suitable strain for fermentation. The effects of PYFS on the cell cycle distribution, growth factors, inflammatory factors of HDPCs, as well as its hair growth-promoting mechanism, were investigated. Experiments revealed that after fermentation, the proportion of cells in the G0/G1 phase decreased by 11.09%, while the proportion of cells in the S phase increased by 35.44%. Additionally, the level of the growth factor VEGF increased by 42.34%, while the level of the inflammatory factor TGF-β1 decreased by 23.81%. Moreover, the fermentation process correlates with altered mRNA expression of Wnt/β-catenin pathway-related genes by upregulating the mRNA expression levels of β-catenin, DVL1, and LEF1, and downregulating the mRNA expression level of DKK-1. Finally, non-targeted metabolomics technology was used to analyze the metabolite changes after fermentation. The most significant differential metabolites mainly include flavonoids, amino acids and their derivatives, and organic acids and their derivatives. This study utilized microbial fermentation technology to prepare the yeast fermentation solution, selected the optimal fermentation strain, and demonstrated that its fermentation product significantly promotes HDPC metabolic activity, supports hair follicle health by regulating the balance of growth factors, alters expression patterns of Wnt/β-catenin pathway-related genes, and substantially alters the metabolite composition of Platycladus orientalis leaves extract through fermentation. Full article

Aiming at the problem that the current traditional apple yield estimation methods rely on manual investigation and do not make full use of multi-source information, this paper proposes an apple orchard yield prediction framework combining multispectral remote sensing features and ground features. The framework is oriented to the demand of yield prediction at different scales. It can not only realize the prediction of apple yield at the district and county scales, but also modify the prediction results of small-scale orchards based on the acquisition of orchard features. The framework consists of three parts, namely, apple orchard planting area extraction, district and county large-scale yield prediction and small-scale orchard yield prediction correction. (1) During apple orchard planting area extraction, the samples of some apple planting areas in the study area were obtained through field investigation, and the orchard and non-orchard areas were classified and discriminated, providing a spatial basis for the collection of subsequent yield prediction-related data. (2) In the large-scale yield prediction of districts and counties, based on the obtained orchard-planting areas, the corresponding multispectral remote sensing features and environmental features were obtained using Google Earth engine platform. In order to avoid the noise interference caused by local pixel differences, the obtained data were median synthesized, and the feature set was constructed by combining the yield and other information. On this basis, the feature set was divided and sent to Apple Orchard Yield Prediction Network (APYieldNet) for training and testing, and the district and county large-scale yield prediction model was obtained. (3) During the part of small-scale orchard yield prediction correction, the optimal model for large-scale yield prediction at the district and county levels is utilized to forecast the yield of the entire planting area and the internal local sampling areas of the small-scale orchard. Within the local sampling areas, the number of fruits is identified through the YOLO-A model, and the actual yield is estimated based on the empirical single fruit weight as a ground feature, which is used to calculate the correction factor. Finally, the proportional correction method is employed to correct the error in the prediction results of the entire small-scale orchard area, thus obtaining a more accurate yield prediction for the small-scale orchard. The experiment showed that (1) the yield prediction model APYieldNet (MAE = 152.68 kg/mu, RMSE = 203.92 kg/mu) proposed in this paper achieved better results than other methods; (2) the proposed YOLO-A model achieves superior detection performance for apple fruits and flowers in complex orchard environments compared to existing methods; (3) in this paper, through the method of proportional correction, the prediction results of APYieldNet for small-scale orchard are closer to the real yield. Full article

Background/Objectives: Adults with intellectual disability (ID) experience high rates of psychiatric comorbidity but often face diagnostic challenges and treatment barriers, leading to inappropriate psychotropic medication use. This study examined the extent to which specialized psychiatric assessment and improved diagnostic accuracy had an impact on medication management and clinical outcomes in adults with ID and co-occurring psychiatric disorders. Methods: This observational retrospective study analyzed medical records from 25 adults with ID who underwent specialized psychiatric assessment at a community-based service in Italy between January 2023 and January 2024. Psychopathological diagnoses were established according to Diagnostic Manual—Intellectual Disability, Second Edition (DM-ID2) criteria, based on clinical observation and a comprehensive assessment using validated instruments. Clinical outcomes were assessed using a psychometric tool encompassing multiple psychopathological and behavioral dimensions. Data on psychotropic prescriptions and side effects were also collected. Non-parametric analyses were performed, with significance set at α = 0.05. Results: The proportion of patients with a psychiatric diagnosis increased from 32% to 96% after specialized assessment (p < 0.001), with notable rises in depressive (0% to 32%), bipolar (8% to 36%), anxiety (4% to 24%), and impulse control (0% to 16%) disorders. First-generation antipsychotic prescriptions decreased (from 36% to 8%, p = 0.023), while antidepressant use increased (from 12% to 52%, p = 0.004). The mean number of side effects per patient declined from 1.6 to 0.5 (p < 0.001), particularly the elevated prolactin level and psychomotor retardation. Significant improvements were observed in symptom intensity and frequency across multiple domains, including aggression, mood disturbances, and compulsions (p < 0.001). Conclusions: In this single-center retrospective study, specialized psychiatric assessment was associated with improved diagnostic accuracy, medication management, and clinical outcomes in adults with ID. The increase in psychiatric diagnoses likely reflects improved identification, addressing key challenges in precision diagnosis for people with neurodevelopmental disorders. Although the overall number of prescribed medications remained stable, optimization of treatment regimens reduced first-generation antipsychotic use and related adverse effects. These findings indicates that access to specialized assessment and precision diagnosis could improve psychopharmacological interventions and outcomes for this vulnerable population, but larger, multi-center and longer-term studies are needed to confirm these results. Full article

Aim: This study aimed to assess self-reported oral and orofacial health, hygiene habits, and oral health knowledge among Croatian athletes, and to determine factors influencing that knowledge. Differences between contact and non-contact sports, as well as the occurrence of dental trauma and temporomandibular joint (TMJ) symptoms, were also examined. Methods: A cross-sectional, questionnaire-based study was conducted among 1007 athletes (56% male, 44% female) aged 18–42 years, recruited through national sports federations and university sports clubs. The instrument comprised 85 items divided into five domains: sociodemographic data, oral hygiene habits, self-assessed oral health, TMJ symptoms, and oral health knowledge. Data were analysed using descriptive statistics, Chi-square and Fisher’s exact tests, and generalised linear modelling (p < 0.05). Results: Athletes demonstrated moderate oral health knowledge (mean score 11.3 ± 4.4/18). While 92.2% recognised that poor oral hygiene leads to caries and periodontitis, only 52.4% correctly identified the ideal time to replant an avulsed tooth. Female participants, older age groups, and those with higher education had significantly better knowledge (p ≤ 0.05). Recreational athletes scored higher than amateurs (p = 0.002), and those with prior dental trauma experience also showed greater awareness (p = 0.028). No significant difference was found between contact and non-contact sports (p = 0.287). Despite good brushing habits (86.9% brushed twice daily), only 25.4% regularly used dental floss or interdental brushes. A small proportion of athletes reported symptoms related to temporomandibular joint function, most commonly joint clicking (18.2%), tooth wear (13.4%), and nocturnal bruxism (14.3%). There were no significant differences between contact and non-contact sports, except for muscle stiffness near the temples (p = 0.024) and daytime or stress-related teeth grinding (p = 0.013 and p = 0.018). Conclusions: Croatian athletes demonstrated moderate oral health knowledge and satisfactory hygiene habits, but preventive practices remain inadequate. Education level, gender, and previous dental trauma were key determinants of knowledge. Systematic preventive programmes and targeted education are necessary to improve oral health awareness in sports populations. Full article

The study focuses on the design of an intelligent information-control system (ICS) for metallurgical production, aimed at robust forecasting of technological parameters and automatic self-adaptation under noise, anomalies, and data drift. The proposed architecture integrates a hybrid LSTM–DNN model with low-discrepancy hypercube sampling using Sobol and Halton sequences to ensure uniform coverage of operating conditions and the hyperparameter space. The processing pipeline includes preprocessing and temporal synchronization of measurements, a parameter identification module, anomaly detection and correction using an ε-threshold scheme, and a decision-making and control loop. In simulation scenarios modeling the dynamics of temperature, pressure, level, and flow (1 min sampling interval, injected anomalies, and measurement noise), the hybrid model outperformed GRU and CNN architectures: a determination coefficient of R² > 0.92 was achieved for key indicators, MAE and RMSE improved by 7–15%, and the proportion of unreliable measurements after correction decreased to <2% (compared with 8–12% without correction). The experiments also demonstrated accelerated adaptation during regime changes. The scientific novelty lies in combining recurrent memory and deep nonlinear approximation with deterministic experimental design in the hypercube of states and hyperparameters, enabling reproducible self-adaptation of the ICS and increased noise robustness without upgrading the measurement hardware. Modern metallurgical information-control systems operate under non-stationary regimes and limited measurement reliability, which reduces the robustness of conventional forecasting and decision-support approaches. To address this issue, a hybrid LSTM–DNN architecture combined with low-discrepancy hypercube probing and anomaly-aware data correction is proposed. The proposed approach is distinguished by the integration of hybrid neural forecasting, deterministic hypercube-based adaptation, and anomaly-aware data correction within a unified information-control loop for non-stationary industrial processes. Full article