Search Results (326)

Background/Objectives: The health status of institutionalized older adults is determined by the interaction of functional, cognitive, nutritional, anthropometric, and biochemical factors, which may not be adequately captured through single-domain assessments. Within the framework of the VIVA Project (Vulnerability Index: Valencia institutionalized Adults), this study aimed to characterize institutionalized older adults during the COVID-19 pandemic using an integrated multidimensional approach and to explore clinically interpretable vulnerability profiles. Methods: This cross-sectional study included 124 residents from 10 nursing homes of Valencia, Spain. Data were obtained from institutional records and included age, sex, body mass index (BMI), Barthel Index, Mini-Examination of Cognition (MEC), Tinetti scale, Mini Nutritional Assessment-Short Form (MNA-SF), and biochemical markers related to protein status, lipid metabolism, micronutrient availability, and renal function. An exploratory VIVA multidimensional index was constructed from nine standardized variables, and k-means clustering was applied to these variables rather than to a single summed score to identify residents’ phenotypes. An exploratory logistic regression model was used to assess the internal discrimination of the high-vulnerability phenotype. Results: The cohort showed marked heterogeneity across functional, cognitive, nutritional, anthropometric, and biochemical domains. Cluster analysis identified three clinically interpretable phenotypes ranging from lower to higher vulnerability. Functional impairment, particularly the Barthel Index and Tinetti score, was the main driver of separation between phenotypes, while biochemical markers contributed to refining profile discrimination. The exploratory logistic regression model showed high internal discrimination for the high-vulnerability phenotype, supporting the internal coherence of the integrated framework. Conclusions: An integrated multidimensional framework may be useful for characterizing vulnerability among institutionalized older adults and supporting risk stratification in long-term care settings. The logistic regression findings, including the high AUC, should be interpreted only as evidence of internal discrimination and internal coherence of the exploratory construct, not as evidence of external validity, reproducibility, diagnostic accuracy, or future predictive utility. Full article

Early fault detection in microsatellite thermal control systems (TCS) faces fundamental challenges: high-dimensional redundant telemetry channels, overlapping multi-scale periodicities that obscure anomaly signatures, and severely limited daily data downlink (1–2 passes per day) that restricts the temporal window for diagnosis. Existing data-driven approaches either rely on supervised learning, requiring labeled fault data that are scarce in practice, or employ univariate analysis that fails to capture inter-sensor spatial correlations. To address these limitations, this paper introduces a hybrid framework integrating information-theoretic feature selection and spatiotemporal deep learning. The Generalized Maximum Information Coefficient (GMIC) quantifies nonlinear dependencies between temperature channels for key channel selection, reducing dimensionality by 82% while preserving diagnostic information. A dual-level Seasonal Trend Decomposition (STL) method disentangles orbital-periodic dynamics from diurnal cycles, effectively isolating distinct thermal characteristics at multiple timescales. Each decomposed component is modeled using Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) networks to capture spatiotemporal dependencies for accurate temperature prediction. An adaptive threshold-based weighted error fusion mechanism enables early fault detection within a single day of telemetry data. Experimental validation on real satellite telemetry data demonstrates that the proposed framework achieves high-precision fault detection across multiple fault types using a minimal set of temperature channels, significantly outperforming existing benchmarks in both prediction accuracy and detection reliability. Full article

Wind turbines are increasingly deployed at larger scales and in harsher operating environments, leading to greater structural complexity, stronger load variability, and higher maintenance demands across both drivetrain and structural components. Reported field data indicate that gearboxes and bearings account for approximately 30–40% of total turbine downtime, while blade-related failures contribute roughly 20–25% of reported failure events, primarily through fatigue, delamination, leading-edge erosion, and lightning-induced defects. In parallel, large-scale and offshore turbines show increasing susceptibility to tower fatigue cracking, corrosion-assisted degradation, and flange joint bolt-preload loss under cyclic and environmental loading. This review provides a comprehensive applied-engineering synthesis of failure mechanisms, reliability challenges, and monitoring strategies for major wind turbine components, including gearboxes, bearings, blades, towers, and flange joints. A wide range of condition monitoring, structural health monitoring (SHM), and prognostics and health management (PHM) approaches is critically examined, including vibration analysis, acoustic emission, ultrasonic inspection, infrared thermography, impedance-based sensing, electromagnetic methods, machine vision, SCADA-based diagnostics, and artificial-intelligence-assisted fault classification. The review compares these techniques in terms of detectable damage types, spatial coverage, sensitivity, deployment practicality, and limitations under real operating conditions. In addition, statistical reliability methods and data-driven models are discussed to interpret failure trends and uncertainty. Recent AI-based studies have reported fault classification accuracies exceeding 90% under controlled or semi-controlled conditions; however, their field reliability remains constrained by data imbalance, domain shift, limited labeled failure datasets, model interpretability, and insufficient validation under realistic turbine operating environments. The main contribution of this review is an integrated applied synthesis that connects drivetrain and structural failure mechanisms with measurable monitoring indicators, diagnostic technologies, AI-enabled PHM limitations, and predictive-maintenance decision needs. The paper provides practical guidance for monitoring design, early fault detection, predictive maintenance, and long-term reliability improvement in next-generation wind turbine systems. Full article

The high prevalence of cardiovascular diseases and the extensive application wearable electrocardiogram (ECG) devices for long-term monitoring have posed significant challenges for the transmission, storage, and real-time processing of massive amounts of ECG data. Consequently, efficient ECG compression and reconstruction have become a research priority in remote ECG monitoring. Traditional compressed sensing is complex and has high computational overhead, while single deep learning models cannot simultaneously extract local waveforms and model temporal dependencies. To address these shortcomings in the reconstruction process, this paper presents a CNN-LSTM-Attention hybrid model. This model utilizes a convolutional neural network (CNN) to capture local ECG waveform features, employs a long short-term memory (LSTM) network to learn long-term temporal dependencies, and introduces an attention mechanism to weight and fuse key diagnostic features, enabling accurate focus on key components including the QRS complex and ST segment. Experimental results on the MIT-BIH Arrhythmia dataset demonstrate that across the full compression range of 0.1–0.9, the proposed model achieves favorable comprehensive performance. Its PRD is stabilized at 10–12%, the SNR stays above 20 dB, and the RMSE is mostly lower than 0.25 mV. In terms of reconstruction accuracy and stability, our model outperforms the single CNN and CNN-LSTM models by a large margin. Full article

Precision agriculture increasingly relies on high-resolution, long-term remote sensing to delineate sub-field management zones. However, traditional spatial zonation assumes temporal stationarity, utilizing seasonal aggregates that obscure transient, intra-annual stress signals. This study develops a data-driven framework to characterize both persistent and non-stationary crop water use dynamics by integrating monthly, 30-m evapotranspiration (ET) data from OpenET (2000–2025) with zero-shot temporal anomaly detection. A pre-trained time-series foundation model (Chronos-T5-Small) generated counterfactual expectations for sub-field ET, quantifying deviations using a mean absolute error-based anomaly score. Unsupervised clustering of these anomaly scores with longitudinal ET metrics partitioned the landscape into dynamic biophysical regimes. Cross-registered against legacy persistence mapping based on seasonal totals, the foundation model showed strong directional agreement (86.1%, Cohen’s Kappa = 0.716) in identifying chronically constrained zones across 869 shared active pixels. Crucially, the framework identified 966 historically persistent pixels undergoing stability decay, of which 95.3% were statistically verified via paired t-tests to have collapsed into the field’s baseline variance pool. Furthermore, counterfactual anomaly detection isolated zones of recent acute divergence, differentiating enduring edaphic constraints from sudden system disruptions. This approach demonstrates how foundation models can transition from purely predictive engines to diagnostic instruments, advancing operational precision agriculture. Full article

Fault detection and diagnosis of three-phase inverter-fed motor drives is essential for ensuring system reliability, safety, and continuous operation in applications such as electric vehicles and industrial automation. This paper proposes a data-driven fault detection framework based on normalized current features and a lightweight bidirectional long short-term memory (BiLSTM) network which can be generalized to different motor power rating in the same controller system. A compact set of six time-domain features, consisting of the mean and root-mean-square (RMS) values of the phase currents, is extracted and normalized with respect to the average RMS value. This normalization effectively removes dependency on operating conditions, enabling the model to generalize across different load levels and motor power ratings without retraining. A lightweight BiLSTM architecture is employed, reducing computational complexity while maintaining high diagnostic performance. The proposed method is validated under various operating conditions, including different speeds, load variations, motor power ratings, and noisy conditions. The results demonstrate an overall classification accuracy of 99.65%, with reliable fault detection achieved within less than half of a fundamental cycle. The proposed approach provides an efficient, robust, and scalable solution for inverter fault detection and diagnosis, offering strong potential for practical deployment in modern motor drive systems. Full article

Background/Objectives: Serum neuron-specific enolase (NSE) is a biomarker of neuronal injury, but its prognostic role in pediatric convulsive status epilepticus (CSE) remains uncertain. We evaluated the association between serum NSE levels and short-term neurological outcome, assessed model calibration with internal bootstrap validation, and examined whether NSE provides incremental discrimination beyond established clinical severity scores. Methods: This was a single-center retrospective cohort study of children aged 1 month to 18 years admitted to a tertiary pediatric intensive care unit (PICU) with CSE as the primary admission diagnosis between January 2024 and November 2025. The primary outcome was poor neurological outcome at hospital discharge, defined as a worsening of ≥1 point in the Pediatric Cerebral Performance Category (PCPC) score from baseline (ΔPCPC ≥ 1) or in-hospital death. A multivariable logistic regression model adjusting for NSE, PRISM III, acute symptomatic etiology, and mechanical ventilation was developed, with bootstrap optimism-corrected internal validation (2000 resamples) and formal calibration assessment. Separate models for in-hospital mortality and for neurological deterioration among survivors were conducted as secondary analyses. Diagnostic operating characteristics were reported with 95% Wilson confidence intervals. The study followed the STROBE and TRIPOD reporting guidelines. Results: Of 132 children included (median age 26 months, 56.1% male), 60 (45.5%) had a poor neurological outcome including 18 deaths (13.6%). Serum NSE was significantly higher in the poor-outcome group (median 22.0 vs. 14.4 μg/L; p < 0.001). In the primary multivariable model, NSE (adjusted OR 1.11 per μg/L; 95% CI 1.06–1.19; p = 0.001) and PRISM III (adjusted OR 1.15; 95% CI 1.03–1.37; p = 0.013) were independently associated with poor outcome. The model showed acceptable calibration (Hosmer–Lemeshow p = 0.130) and a bootstrap optimism-corrected AUC of 0.759. NSE remained independently associated with both in-hospital mortality (aOR 1.13) and with ΔPCPC ≥ 1 in survivors (aOR 1.09). The AUC for NSE alone was 0.741 (95% CI 0.65–0.82) for poor outcome and 0.885 (0.79–0.96) for mortality. The combined PRISM III + NSE model showed a numerically higher but not statistically significant AUC compared with PRISM III alone (0.784 vs. 0.726; DeLong p = 0.103). Conclusions: Higher serum NSE is independently associated with adverse short-term neurological outcome and mortality in pediatric CSE, including in survivor-only analysis. However, the present data do not demonstrate clinically meaningful incremental prognostic value beyond PRISM III, and the proposed cutoff was derived and tested in the same cohort and is therefore optimistic. These findings are hypothesis-generating and require external validation in prospective multicenter cohorts with serial sampling and long-term neurodevelopmental follow-up before routine clinical use can be advocated. Full article

Background: Sepsis remains a leading cause of morbidity, mortality, and healthcare expenditure worldwide. Despite international guidelines and diagnostic criteria, real-world variability in coding, treatment, and outcomes persist. This retrospective study analyzed 13,994 coded sepsis-related hospitalizations identified through administrative ICD-9-CM algorithms between 2016 and 2024 to evaluate the burden of sepsis, temporal trends, clinical outcomes, and healthcare costs within a regional health system. Methods: Hospitalization data across four local health authorities (ASL 201–204) over an 8-year period were analyzed. The coded sepsis cases were identified using validated ICD-9-CM-based algorithms and classified into four groups according to available microbiological coding: Gram-positive, Gram-negative, anaerobic and unspecified. Variables included patient demographics, length of stay, costs, outcomes (in-hospital and post-discharge mortality) and presence of septic shock. Comparative analyses were conducted using descriptive statistical methods and One-way ANOVA test and chi-squared tests were applied to evaluate the significance of differences. Multivariable logistic regression models were used to identify independent predictors of 6- and 12-month mortality. Results: The dataset included 13,994 coded sepsis-related hospitalizations, with the largest subgroup being ‘unspecified’ (48.0%). Among cases with specified etiology, coded anaerobic sepsis categories, though rare (0.7%), were associated with higher in-hospital mortality (45.5%) and economic burden (avg. € 8563). Mortality remained high at 6 and 12 months across all types, exceeding 50% post-discharge. Increasing age (OR ≈ 1.06 per year) and septic shock (OR ≈ 4.5–4.8) were the strongest independent predictors of mortality. Differences across microbiological groups should be interpreted cautiously given the high proportion of cases without organism-specific coding. Despite a modest reduction in mortality over time, sepsis was associated with persistently high 6- and 12-month mortality, highlighting a substantial long-term burden beyond the acute phase of illness. These findings suggest that sepsis-related hospitalizations are associated with substantial long-term mortality beyond the acute phase of illness. Discussion: These findings underscore the clinical and economic impact of sepsis in hospitalized patients, across microbiological coding categories. The high mortality rate at 6–12 months may support the need for further investigation into structured post-discharge follow-up strategies. Sepsis represents a substantial clinical and economic burden within the regional healthcare system, with persistently elevated short- and mid-term mortality. Incomplete organism-level documentation limits direct etiologic comparisons and highlights the need for improved integration between clinical, microbiological, and administrative data systems. Future research should integrate clinical variables and lab results to enable risk stratification and intervention planning. Full article

Background/Objectives: Type 2 diabetes mellitus (T2DM) is a prevalent metabolic disorder associated with substantial long-term morbidity and mortality. Routinely collected anthropometric, biochemical, and hematological variables may contain useful discriminatory information for data-driven classification. This study aimed to compare the apparent classification performance of multiple machine learning algorithms for distinguishing individuals with and without T2DM using routinely obtained clinical parameters in a single-center dataset. Methods: This single-center observational study included 160 adults (95 females, 65 males) evaluated at the Endocrinology Outpatient Clinic of Gaziantep Islam Science and Technology University, Faculty of Medicine, Ersin Arslan Training and Research Hospital. The dataset comprised anthropometric measurements, biochemical markers, and complete blood count parameters. SMOTE was applied only within the training folds to address class imbalance and to avoid information leakage. Following fold-internal data preprocessing, which included imputing missing values and feature standardization where appropriate, the dataset was evaluated using stratified 5-fold cross-validation. SHAP analysis was performed to interpret the model predictions. A calibration curve was used to assess the model’s reliability. Eight supervised machine learning models were evaluated with and without HbA1c: Logistic Regression, Linear Discriminant Analysis, Quadratic Discriminant Analysis, Decision Tree, Random Forest, Extra Trees, Gaussian Naive Bayes, and k-Nearest Neighbors. Model performance was evaluated using accuracy, sensitivity, specificity, and F1 score, and ROC curves were used as a diagnostic tool. Results: The models were evaluated in two different ways: with and without HbA1c. Random Forest demonstrated the best classification performance in the cross-validated evaluation; without HbA1c, it achieved 92.2% accuracy, 93.9% sensitivity, 97.9% specificity, and a 95.9% F1 score. When HbA1c was included, it achieved 98.0% accuracy, 97.9% sensitivity, 98.8% specificity, and a 99.0% F1 score. Decision Tree and Extra Trees demonstrated strong performance with accuracy rates of 87.6% and 92.8%, respectively, without HbA1c, and 90% and 93.5% when HbA1c was included; in contrast, KNN yielded the lowest accuracy rate (70.6%). Overall, tree-based models performed better than linear classifiers on this dataset. Conclusions: Machine learning models based on routine clinical and anthropometric variables demonstrated promising performance for T2DM classification in this single-center dataset; tree-based approaches yielded the most promising results. Including HbA1c improved the models’ ability to classify individuals with and without T2DM. However, since HbA1c was included both as a predictor and as part of the operational definition of the diabetes group, the findings should be interpreted with caution due to the risk of target leakage. Therefore, these results should be considered exploratory rather than evidence of clinically applicable predictive performance, and an independent external validation study should be conducted prior to clinical application. Full article

Geometric monitoring of slender support structures, particularly steel lattice transmission towers, is a critical component of power infrastructure diagnostics. Such structures are susceptible to environmental influences and long-term deformation processes, which necessitates precise assessment of their geometric axis. The aim of this study was to evaluate the influence of the terrestrial laser scanning (TLS) scanner class and point cloud registration strategy on the determination of the geometric axis of a steel high-voltage lattice transmission tower (hereafter LTT). Unlike previous studies focused primarily on TLS-based axis reconstruction, this work introduces a comparative assessment of registration strategies, an error propagation model, and the proposed Axis Drift Index (ADI) as quantitative indicators of axis stability. The analysis was based on data obtained using a tachymetric method (reference), a compact scanner (Leica BLK360), and a survey-grade scanner (Riegl VZ-400i). The comparison included planimetric axis deviation, consistency of deformation direction, variation in results with height, and the influence of registration quality. The results show that TLS measurements performed using a survey-grade scanner and target-based registration exhibit high agreement with tachymetric results. In contrast, cloud-to-cloud registration without a stable reference framework leads to cumulative errors and instability of the reconstructed axis, particularly in the upper parts of the structure. The observed deviations in the BLK360 dataset were dominated by registration-related geometric instability rather than unequivocal structural deformation signals. The findings indicate that the accuracy of geometric axis determination in slender structures is governed more by the adopted point cloud registration strategy than by the scanner class itself. The proposed ADI parameter and linear error propagation model additionally enabled a quantitative assessment of geometric consistency with height. From an engineering perspective, this highlights the importance of stable reference systems and appropriate survey design in high-precision TLS applications. Although the study was conducted on a single lattice tower, the results provide practical insight into the reliability of TLS workflows for slender structures characterized by discontinuous geometry and high sensitivity to registration errors. Full article

Background: Long-duration spaceflight (LDSF) poses unique challenges to ocular health as microgravity, radiation, and environmental changes can cause lasting visual and structural impairments that affect astronaut performance. Objective: This review synthesises current evidence on in- and post-flight ocular complications. It integrates clinical findings, terrestrial analogues, animal studies, and theoretical models to characterise the pathophysiology, risk factors, and countermeasures associated with spaceflight-induced ocular changes. Methods: A review of peer-reviewed literature was conducted, focusing on dry eye disease, corneal edema, ocular biometric shifts, spaceflight associated neuro-ocular syndrome (SANS), and radiation-induced cataractogenesis. Data from in-flight imaging, post-flight assessments, and ground-based analogues were analysed. Results: Spaceflight induces multifactorial ocular changes, including tear film instability, optic disc edema, posterior globe flattening, and hyperopic refractive shifts. These effects are thought to result from cephalad fluid shifts compartmentalised cerebrospinal fluid pressure, venous congestion, and impaired glymphatic system. Long-term risks, such as cataractogenesis, are linked to radiation exposure and genetic susceptibility. Although several countermeasures are being explored, no single approach fully prevents these complications. Conclusions: Ocular complications during LDSF remain a significant challenge for astronaut health and mission performance. A multimodal approach combining mechanical, nutritional, and diagnostic strategies will be essential for future exploration-class missions. Further research is needed to refine countermeasures and preserve astronauts’ visual function. Full article

Failures in critical systems and equipment within nuclear power plants (NPPs) significantly threaten operational safety and reliability. Therefore, rapid and accurate root cause localization during the incipient stages of failure is critical to preventing escalation. Traditional modeling methods often fail to address the inherent structural complexity of NPPs, the diversity of failure modes, and the stochastic mapping relationships between symptoms and causes. To address these challenges, this paper proposes an intelligent fault diagnosis framework integrating knowledge graphs (KGs) and Bayesian networks (BNs). First, by analyzing failure modes and anomaly characteristics, we define discrimination criteria for typical faults. Second, a structured knowledge modeling approach is developed to transform unstructured fault information into a KG, which is subsequently mapped to a BN topology. Finally, to mitigate the subjectivity of expert priors, data-driven structure and parameter learning algorithms are employed to optimize the model, enhancing inference accuracy. Robustness was validated through experiments targeting three fault severity levels, using signed directed graphs (SDGs), support vector machines (SVMs), domain generalization softmax (DG-softmax) and long short-term memory (LSTM) as benchmarks. Experimental results demonstrate that the proposed method maintains high diagnostic precision across varying severities, outperforming traditional data-driven methods in accuracy and stability. This study enhances the interpretability and engineering applicability of intelligent diagnosis in nuclear power systems. Full article

The contact performance of automotive airbag electrical connectors directly affects the stable conduction of the initiator circuit, yet sufficient failure data are difficult to obtain for such long-life safety-critical components. This study develops a degradation model for connectors with stainless-steel pins, beryllium-bronze sockets, and Ni/Au composite coatings, using the contact resistance increment as the degradation measure. Considering the accumulation of oxidation corrosion products under thermal stress, as well as the local film rupture and re-oxidation induced by fretting wear under combined temperature-vibration stress, a nonlinear time scale t^α is introduced to describe the nonlinear growth of contact resistance. A random-drift nonlinear Wiener process is then constructed: the diffusion term represents local fluctuations within each sample trajectory, while the random drift rate captures growth-rate differences among samples. Parameter estimation was performed using degradation data obtained from 160 °C high-temperature and 160 °C temperature-vibration accelerated degradation tests. The estimation results show that the stress-class-specific time-scale model better reflects the different degradation mechanisms than a common time-scale model, and that the temperature-vibration group exhibits higher resistance growth and stronger trajectory fluctuations. Model diagnostics support the description of the main increment distribution and sample-to-sample differences, while EDS and XPS results provide supplementary evidence for oxidation-related surface composition changes and coating-state evolution. Full article

Background and Objectives: Despite global commitment to the World Health Organization triple elimination initiative, evidence on integrated antenatal service delivery models that simultaneously address human immunodeficiency virus (HIV), syphilis, and hepatitis B virus (HBV) remains fragmented, particularly across diverse health-system contexts. Eliminating vertical transmission of HIV, syphilis, and HBV is a global priority. Pregnant women are disproportionately affected by these infections, and untreated maternal disease leads to significant infant morbidity. Integrating antenatal screening and treatment provides an opportunity to address all three conditions simultaneously. Purpose: This systematic review and meta-analysis aimed to identify and synthesise evidence on integrated antenatal service delivery models addressing HIV, syphilis, and HBV simultaneously within maternal health services. It specifically examined model characteristics, screening uptake, treatment and follow-up outcomes, implementation barriers and facilitators, and evidence on cost-effectiveness. Methods: This systematic review and meta-analysis followed PRISMA 2020 guidelines and was registered in PROSPERO (CRD420261342186). We searched Scopus, PubMed, Web of Science, and Dimensions for studies published between January 2007 and January 2026. Of 423 records identified, 11 met the inclusion criteria after excluding two studies that did not provide empirical results for an integrated service model addressing all three target infections simultaneously. Data on study characteristics, service delivery, diagnostics, outcomes, and implementation factors were extracted. A random-effects meta-analysis of proportions was conducted using the DerSimonian–Laird estimator with logit transformation. Results: Eleven studies covered Asia, Africa, Europe, and Latin America, mostly in low- and lower-middle-income countries. Integration ranged from rapid test packages in community clinics to comprehensive programmes including STI treatment, malaria testing, and HBV birth-dose vaccination. Pooled triple testing uptake was 97% (95% CI 92 to 100%). Large programmes achieved over 99% coverage and reduced HIV vertical transmission to below 3%. Pilot studies showed feasibility but noted stockouts, data gaps, and weak treatment linkage. Economic analyses supported cost-effectiveness. Conclusions: Integrated antenatal services appear feasible and can achieve high testing uptake, particularly in well-supported programmes. However, evidence remains uneven regarding treatment completion, infant follow-up, HBV prophylaxis, long-term transmission outcomes, and sustainability in resource-constrained settings. Key challenges include supply constraints, workforce limitations, and follow-up gaps. Future research should evaluate the full care cascade, not screening uptake alone. Full article

Underground space is increasingly positioned in contemporary urban discourse as a strategic resource for sustainable spatial and territorial development, particularly under conditions of limited surface capacity, growing infrastructural demand, and the need for long-term urban resilience. However, its implementation remains constrained by insufficient institutional, planning, and governance integration. Starting from this problem, this paper assesses the institutional readiness of Serbia’s spatial and urban planning system for the integration of underground planning into the territorial development system. The methodological approach is based on the development of an analytical framework for institutional readiness, structured around three key dimensions: regulatory–institutional, spatial–infrastructural, and governance–coordination. This research is conducted through a qualitative analysis of legislative, strategic, planning, and supplementary sources, using stratified criteria—normative, operational, and integrative levels—which enables a structured, document-based diagnostic assessment of the current state of the system. The results indicate that institutional readiness in Serbia is at a low to medium-low level. Although a partially developed normative framework and certain technical-informational capacities exist, underground space is not clearly recognised as a distinct planning category or as an integrated three-dimensional spatial resource. The spatial–infrastructural dimension reveals the existence of relevant cadastral, geospatial, and infrastructural foundations, but without their sufficient integration into a unified 3D planning and governance system. The key limitation is identified in the governance–coordination dimension, where fragmented competences, uneven local capacities, and the absence of dedicated coordination mechanisms hinder the systematic application of underground planning. The paper concludes that the integration of underground planning in Serbia requires gradual institutional transformation toward an integrated, three-dimensional, and long-term-oriented model of spatial governance. Its contribution lies in formulating an initial diagnostic framework that connects debates on planning systems, institutional fragmentation, spatial data integration, and territorial governance, and may serve as a basis for further research and policy development in the field of integrated territorial development. Full article