Search Results (516)

Monitoring surface subsidence in mining areas is essential for geological disaster early warning and safe production. Existing geometric difference methods heavily rely on the local consistency of multi-temporal point clouds. When horizontal displacement and vertical subsidence are coupled, horizontal movements often cause local misalignments, leading to spatial deviations and discrete anomalies in vertical estimations. To address this issue, this paper proposes DL-C2C, a deep learning model for subsidence extraction from bi-temporal ground point clouds. Within a unified framework, the model introduces horizontal displacement as an auxiliary constraint into the vertical solving process, effectively improving the stability of vertical subsidence estimation through continuous cross-temporal alignment and correlation updating. For feature extraction, DL-C2C employs a PointConv multi-scale pyramid combined with a proposed scale-adaptive Transformer to enhance cross-scale information interaction under sparse and non-uniform sampling conditions. Furthermore, the network constructs dynamic local associations through iterative alignment within a recursive framework, and introduces diffusion-based residual correction at the fine-scale stage to compensate for detail errors at subsidence basin boundaries and in data-missing regions. Experiments on simulated and real-world datasets—covering aeolian sand and mountainous gully landforms—demonstrate that the method achieves mining 3D error (M3DE) of 0.16 cm and 0.22 cm in simulated scenarios. In real-world mining area validations, compared to existing methods, DL-C2C significantly reduces discrete anomalous points, yields an error distribution closer to zero, and exhibits superior performance in boundary transition continuity and non-subsidence area stability. In conclusion, this model provides reliable technical support for large-scale, high-precision intelligent monitoring of geological disasters in mining areas. Full article

Intensifying hydroclimatic extremes have heightened the need for basin-scale indicators of antecedent wetness that are relevant to flood responses. However, ground-based soil-moisture observations are spatially sparse, and satellite products frequently exhibit temporal gaps. To address this limitation, this study integrated satellite- and ground-soil-moisture observations, hydro-meteorological variables, and observed streamflow data from 2018 to 2024 across 26 standard basins (SBs) within three dam basin regions in South Korea: the Nam River Dam (NGD) and the upstream and downstream regions of the Seomjin River Dam (SJD). Using this integrated dataset, we quantified the relationships among precipitation, basin wetness, and rapid discharge increases, subsequently deriving composite thresholds for flood early warnings. For each SB, we trained a Random Forest regression model using satellite-soil-moisture and basin-representative hydro-meteorological inputs—including 1-day accumulated precipitation (P_1d), 7-day accumulated precipitation (P_7d), the antecedent precipitation index (API), and related meteorological variables—to estimate a continuous, daily basin-representative soil-moisture series (SM_RF). Validation results indicated that the coefficient of determination (R²) ranged from 0.6 to 0.7 for most SBs. Extreme event days were consistently associated with elevated values of SM_RF, P_1d, P_7d, and API, demonstrating that antecedent wetness significantly influences the likelihood of rapid discharge events. Finally, composite threshold scanning yielded candidate rules characterized by high precision, moderate hit rates, and low false-alarm rates, confirming the efficacy of the proposed framework for developing flash-flood early-warning thresholds in South Korean dam basins. Full article

Citrus trees exhibit severe alternate bearing, resulting in significant annual yield fluctuations and posing substantial challenges to orchard management planning. Accurate citrus bud counting provides an effective solution by supplying essential data for tree-level and orchard-level yield prediction. However, citrus buds are extremely small (5–10 mm in diameter) and are frequently occluded by leaves during the flowering stage, which makes precise detection highly challenging in complex orchard environments. To address these challenges, this paper proposes a Contextual and Frequency-Aware Detector (CFADet) for robust citrus bud detection. Specifically, an Enhanced Feature Fusion (EFF) module is introduced in the neck to refine multi-scale feature aggregation and strengthen information flow for small targets. A Contextual Boundary Enhancement Module (CBEM) is designed to capture surrounding contextual cues and enhance boundary representation through dimensional interaction and max-pooling operations. To suppress background interference, a Frequency-Aware Module (FAM) is developed to adaptively recalibrate frequency components in the amplitude spectrum, thereby enhancing target features while reducing background noise. In addition, Spatial-to-Depth Convolution (SPDConv) is employed to reconstruct the backbone to preserve fine-grained bud features while reducing model parameters. Experimental results show that CFADet achieves 81.1% precision, 80.9% recall, 81.0% F1-score, and 87.8% mAP, with stable real-time performance on mobile devices in practical orchard scenarios. This study presents a preliminary investigation into robust citrus bud detection in real-world orchard environments and provides a promising technical foundation for intelligent orchard monitoring and early yield estimation, while further validation on larger and more diverse datasets is still required. Full article

Climate extremes are major constraints on agricultural productivity, especially in tropical regions experiencing rapid expansion and intensification of soybean agriculture. This study analyzes spatiotemporal changes in soybean yields in response to droughts and heatwaves across highly productive municipalities in Brazil’s five macroregions from 1989 to 2020. By combining high-resolution meteorological data, satellite-based evapotranspiration estimates, and municipal-level crop yield data, we used standardized drought indices (Standardized Precipitation Index [SPI], Standardized Precipitation Evapotranspiration Index [SPEI]) and a heat index (Warm Spell Duration Index [WSDI]) with spatiotemporal linear regression analyses to explore the links between climate variability and soybean yields across Brazil’s diverse agroclimatic zones. The results show a clear rise in the frequency and severity of compound drought–heat events, especially in the Northeast and South frontiers, where yield sensitivity to hydroclimatic stress is highest. Municipal-level linear regression analyses and spatial patterns indicate that short-term dry events, rather than long-term climate trends, are the main drivers of recent yield variability, with notable spatial spillover effects observed across municipalities. Cristalina and Bom Jesus, for example, exhibit significant negative trends (p < 0.05) in both SPEI-6 (−0.04 and −0.03) and SPI-6 (0.04 and −0.03), indicating a consistent drying tendency over time. Over the 30-year period, municipalities accumulated total soybean yield losses of 3292.3 thousand tonnes (kt), corresponding to an average reduction of 3.7% relative to 5-year detrended yield. These findings highlight the increasing vulnerability of rainfed agriculture in Brazil and emphasize the critical role of seasonal timing, crop phenology, and regional climate patterns for effective climate risk management. This study provides empirical evidence linking combined extremes to agricultural performance and presents a scalable framework for early warning systems and for climate-resilient policy development. Full article

Leaf Area Index (LAI) is critical for monitoring cassava growth and yield prediction, yet ground measurements are time-consuming and labor-intensive for large-scale applications. While satellite-based vegetation indices (VIs) offer a scalable alternative, their performance for cassava LAI estimation remains poorly documented, and optimal index selection for different growth stages is unclear. This study evaluated the predictive performance of 13 Sentinel-2-derived VIs for estimating ground-measured LAI across cassava growth stages. Ground-LAI was measured monthly using a SunScan Canopy Analyzer from January to June 2022 (2–7 months after planting; MAP) in 47 cassava plots in Nakhon Ratchasima Province, Thailand. Linear mixed-effects models and stage-specific regressions assessed VI predictive performance using Coefficient of determination (R²) and Root Mean Squared Error (RMSE). The Green Normalized Difference Vegetation Index (GNDVI) and Normalized Difference Water Index (NDWI) demonstrated superior performance across all growth stages (R² = 0.524; RMSE = 0.350), followed by Sentinel-2 LAI Green Index (SeLI R² = 0.521, RMSE = 0.357). Stage-specific analysis revealed that Ratio Vegetation Index performed best during early growth (2 MAP, R² = 0.671; RMSE = 0.164) while GNDVI and NDWI excelled during mid-growth (3–5 MAP) and SeLI at late growth (7 MAP, R² = 0.393; RMSE = 0.422). While the presence of large trees altered the ranking of VI predictive performance, it did not substantially affect estimation errors, suggesting a relatively small impact of spatial heterogeneity on LAI estimation accuracy. These findings identify GNDVI and NDWI as the most operationally suitable Sentinel-2 indices for cassava LAI estimation and demonstrate that stage-specific index selection can improve monitoring accuracy, providing validated tools for regional-scale cassava crop monitoring using freely available satellite data. Full article

Background: Trapeziometacarpal (TM) osteoarthritis (OA) is a common condition, especially among postmenopausal women, often requiring surgical intervention when conservative treatment fails. In recent years, dual-mobility prostheses have been increasingly used as an alternative to traditional trapeziectomy with suspension arthroplasty. However, limited data exist regarding their comparative cost-effectiveness in public healthcare systems. Purpose: The aim of this study was to compare the cost–benefit ratio and clinical outcomes of two surgical techniques for TM OA: trapeziectomy with suspension arthroplasty and total joint arthroplasty with a dual-mobility prosthesis. Methods: We conducted a retrospective cohort study of 116 hands treated between 2020 and 2024. Patients were divided into two groups based on the surgery they received: trapeziectomy with suspension arthroplasty or implantation of a dual-mobility TM prosthesis. Clinical outcomes were assessed using VAS, DASH, Kapandji score, grip strength, and pinch strength at 12, 36, and 48 months postoperatively. A cost analysis was performed based on hospital reimbursement (Diagnosis-Related Group) and estimated productivity loss. Results: Both techniques yielded significant improvements in pain and function. Patients who were operated on with a prosthesis showed faster recovery and better early outcomes, while the trapeziectomy group had lower direct surgical costs and fewer complications. At 48 months, clinical scores were comparable. The overall cost–benefit ratio favoured trapeziectomy with suspension arthroplasty, while TM prosthesis’s higher costs were justified due to improved short-term functional recovery. Conclusions: Both surgical techniques achieved satisfactory long-term clinical outcomes. The prosthetic option allows for quicker recovery and reduces indirect social costs, while suspension arthroplasty remains more cost-effective for direct costs. These findings highlight the importance of balancing clinical benefit and economic sustainability in surgical decision-making for TM osteoarthritis. Level of Evidence: Level III, retrospective comparative study. Full article

Olive groves in Mediterranean regions are being increasingly exposed to drought and heat extremes, intensifying the interannual yield variability. This study presents an integrated smart-farming framework that links soil context, climate forcing and satellite-observed canopy dynamics to enhance the interpretability and transferability of yield indicators at the parcel scale in southern Spain. Using SoilGrids root-zone properties and the Sentinel-2 time series of the normalized difference vegetation index (NDVI), we first classified parcels into three edaphic clusters. The canopy development was then expressed in thermal time using growing degree days (GDD), enabling phenology-aligned comparisons across campaigns. Two robust patterns emerged: (i) the cumulative NDVI up to 520 GDD showed a consistent negative association with both the biomass and the oil yield, suggesting an early-season vegetation trade-off and carry-over effects typical of perennial systems, and (ii) the rainfall accumulated during a thermally defined window (120–480 GDD) strongly estimated the yield in the subsequent year ($R^2=0.83$–0.97 across soil clusters). By anchoring both vegetation and precipitation indicators to physiologically meaningful thermal milestones, the proposed framework avoids arbitrary calendar windows and enhances the interpretability, cross-year comparability, and scalability. Under projected increases in drought frequency and heat extremes, such hydro-thermal scaling approaches offer a robust basis for early yield forecasting, cooperative-level production planning, and adaptive management in Mediterranean olive systems. Full article

The existing fruit tree yield prediction methods mainly rely on fruit period images or long-term meteorological and soil data, which make it difficult to meet the needs of early yield prediction. In addition, the flowering period images contain complex spatial distribution and severe overlap between flowers, which makes it challenging to directly extract stable structural indicators related to yield. Most existing research has focused on simple statistical indicators such as the number of flowers, while the spatial clustering structure of flowers and their relationship with yield have not been fully explored. Therefore, this article proposes an early apple yield prediction based on flowering stage image thinning simulation characteristics. In this study, blossom images and fruit maturity yield data from 100 apple trees were collected, with flower mask images extracted through standardized image processing. First, the traditional DBSCAN clustering algorithm was enhanced by integrating a KDTree acceleration structure and an adaptive multi-scale mechanism, forming the adaptive multi-scale clustering algorithm (AMS-DBSCAN) to achieve efficient identification of flower clusters and individual flowers. Based on this, two flower thinning simulation strategies based on density and spatial uniformity were designed to model artificial thinning rules and construct multi-dimensional, interpretable phenotypic features. Then, the original statistical features were fused with strategy-generated features and optimized using Lasso. We compared multiple models including XGBoost, BPNN, and SVR for yield prediction. The experimental results showed that XGBoost achieved good predictive performance under the hybrid feature set (R² = 0.856, RMSE = 3.098), which was further improved to R² = 0.900 after feature optimization with Lasso. The results demonstrate that the proposed method enables reliable early yield estimation, providing a new reference for precision management and early decision-making in fruit tree cultivation. Full article

The European Union aims to cut greenhouse gas emissions by 55% by 2030 and reach climate neutrality by 2050, targets that depend on expanding renewable generation in the European energy system. While photovoltaic (PV) capacity has grown quickly in several member states, others remain far behind. This paper frames that divergence as a systems planning problem: installed MW expands through diffusion-like dynamics, but the conversion of investment into energizable capacity is filtered by grid-integration constraints and institutional throughput. The study develops a macro-level framework for systems-level assessment and decision support to guide PV capacity planning and budget allocation using official 2012–2022 data for 22 EU countries. We combine (i) unsupervised clustering of standardized national deployment trajectories, (ii) bounded logistic fits interpreted as an operational diffusion-with-saturation representation that yield comparable growth parameters and maturity years (80–90% of the estimated ceiling), and (iii) a proportional reallocation scenario for countries below 5 GW in 2022. Three clusters emerge—steady growth, early plateau, and atypical paths—and an analytically tractable maturity indicator integrates capacity, rate, and timing in a single measure. In a 10 GW reallocation scenario, average progress toward the 5 GW benchmark rises from 9.8% to 23.1%, closing about 14.8% of the aggregate shortfall. The allocation experiment reveals a clear asymmetry: systems with an existing installed base convert additional MW into benchmark progress more efficiently than very low-baseline systems, where binding constraints are more likely to sit in permitting, interconnection queues, and hosting capacity rather than in finance alone. Turning these allocations into usable capacity depends on timely interconnection and power-electronics integration and on grid-enablement constraints such as interconnection readiness, inverter compliance, and local hosting capacity in high-penetration areas. The contribution is a transparent, updateable decision-support pipeline that links observed trajectory regimes to a maturity “clock” and an auditable allocation baseline, making the trade-off between closing capacity gaps and respecting feasibility filters explicit in an EU system with heterogeneous national subsystems. The proposed approach links macro-level maturity clusters to operational feasibility signals in the grid integration layer, showing that modeling-based allocation can improve system progress but cannot substitute grid-enablement measures, highlighting the importance of regional coordination in the EU energy system under heterogeneous national trajectories. Full article

Objective: This prospective study investigated the use of H-scan ultrasound (US) imaging as a novel component of a multiparametric radiomic analysis framework for characterizing human breast cancer response to neoadjuvant chemotherapy (NAC) before and early after treatment initiation. Methods: Thirty breast cancer patients scheduled for NAC were scanned using a clinical US system (Logiq E9, GE HealthCare) equipped with a 9L-D linear array transducer. Radiofrequency (RF) data was obtained at baseline (pre-NAC) and after 10% and 30% of the complete dose of chemotherapy. The RF data was analyzed by a bank of 256 frequency-shifted bandpass filters to form H-scan US frequency images. Grayscale texture features were extracted from both B-scan and H-scan US images. In addition, US attenuation coefficient and speckle statistics based on the Nakagami and Burr distributions were estimated from the RF data. Data classification of tumor and peri-tumoral regions was performed using a novel three-dimensional (3D) score map based on support vector machine (SVM) modeling. Unlike conventional classifiers that report only a single prediction score, a 3D score map provides a visual representation of the classifier decision space, enabling interpretation of class separation and treatment-induced shifts in multiparametric US measurements. Results: The dataset was split into 10 disjoint partitions (90% training, 10% testing) to compute area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy measures. Actual patient response to NAC was assessed at surgery and categorized as either pathologic complete response (pCR) or non-pCR. Multiparametric US and data classification results at pre-NAC found AUC values of 0.78 after using only tumor information (p < 0.01), which increased to 0.81 with inclusion of peri-tumoral information (p < 0.01). Significant differences in multiparametric US measures from both cancer response types was found after integration of patient data collected at 10% completion of the NAC regimen (i.e., first NAC cycle), yielding an improved AUC of 0.86 (p < 0.001). Conclusions: Multiparametric US imaging with radiomic features from both the tumor and peri-tumoral regions is a promising noninvasive approach for monitoring early breast cancer response to NAC. Full article

This study evaluated the potential of Sentinel-2-derived NDVI (Normalized Difference Vegetation Index) for predicting within-field variability of winter wheat grain yield in central Lithuania during the 2024 growing season. Reliable within-field yield prediction remains challenging in regions with heterogeneous soils and limited region-specific models, particularly in northeastern Europe. Grain yield data were obtained from combine harvesters equipped with GPS yield monitoring across 13 fields with a total area of 283.6 ha. NDVI values were calculated for four half-monthly periods from March to May, corresponding to key phenological stages (from tillering to spike emergence). Spatial and temporal variability in NDVI–yield relationships was observed, with early May consistently showing the strongest correlations (r up to 0.49), particularly in lower-fertility fields, indicating its critical role in yield prediction. Machine learning models (Random Forest, XGBoost, and Deep Neural Networks), along with linear regression, were applied to predict yields based on NDVI from four growth stages. Random Forest achieved the highest predictive accuracy (MAE = 0.951 t/ha), outperforming the other models. The model also showed the highest correlation with observed yields (Pearson r = 0.717), indicating strong agreement between predicted and measured values. Feature importance analysis confirmed NDVI from 1 to 15 May as the most influential predictor across all models. Predicted yield maps closely matched observed patterns, with the largest discrepancies near field edges due to combine harvester effects. These findings highlight the utility of mid-season NDVI for precise estimation of within-field grain yield variability and demonstrate that Random Forest models can effectively capture the NDVI–yield relationship, particularly under heterogeneous field conditions. Full article

Background: Overweight and obesity are emerging public health concerns among young adults. However, evidence on associated sociodemographic, dietary, and lifestyle factors among young industrial workers in low- and middle-income countries remains limited. This study aimed to identify factors associated with overweight and obesity among Vietnamese young industrial workers aged 18–30 years. Methods: A cross-sectional study was conducted among 2295 young industrial workers (55.6% men and 44.4% women) recruited from factories and industrial zones in three geographic regions of Vietnam. Sociodemographic characteristics, dietary habits, lifestyle behaviors, and physical activity were assessed using a structured questionnaire. Body mass index (BMI) was calculated from self-reported height and weight and classified using WHO Western Pacific Region (WPRO) cut-offs; overweight/obesity was defined as BMI ≥ 23.0 kg/m². Physical activity was assessed using the International Physical Activity Questionnaire—Long Form (IPAQ-LF) and categorized by total MET-min/week according to IPAQ scoring guidelines. Logistic regression analyses were performed to estimate crude and adjusted odds ratios (ORs) with 95% confidence intervals (CIs). Results: Overall, 10.4% of participants were overweight (BMI 23.0–24.9 kg/m²) and 7.0% were obese (BMI ≥ 25.0 kg/m²), yielding a combined prevalence of 17.4%. After multivariable adjustment, increasing age (aOR = 1.15; 95% CI: 1.10–1.20), male sex (aOR = 2.10; 95% CI: 1.59–2.76), and regular alcohol consumption (aOR = 1.37; 95% CI: 1.04–1.81) were independently associated with higher odds of overweight/obesity, while residence in the Southern region was inversely associated (aOR = 0.57; 95% CI: 0.42–0.76). High total physical activity (vs. low activity) was positively associated with overweight/obesity, whereas moderate physical activity was not independently associated. Other dietary behaviors were not significantly associated after adjustment. Conclusions: Among Vietnamese young industrial workers, overweight and obesity were associated with age, sex, alcohol consumption, and geographic region. The observed association with high total physical activity likely reflects the occupational context of physical activity in this population, highlighting the importance of distinguishing between occupational and leisure time physical activity when interpreting physical activity obesity relationships. These findings underscore the relevance of early, workplace relevant prevention strategies targeting modifiable behaviors, particularly alcohol use. Full article

Background/Objectives: Breast cancer awareness is essential for early detection and timely help-seeking among women and represents a key component of multidisciplinary breast cancer prevention. The Breast Cancer Awareness Measure (Breast-CAM) is widely used to assess awareness of breast cancer symptoms, risk factors, and screening behaviors. Its measurement quality across populations has not yet been comprehensively evaluated. As Breast-CAM is a population-reported measurement instrument, evaluation using a standardized framework for measurement properties is required. This systematic review and meta-analysis aimed to assess the psychometric properties of the Breast-CAM across diverse populations and cultural adaptations, in accordance with COSMIN (COnsensus-based Standards for the selection of health Measurement INstruments) methodological standards. Methods: Major bibliographic databases and trial registries were systematically searched for peer-reviewed English-language studies published between 2010 and 2025 that evaluated at least one psychometric property of the Breast-CAM in adult women. Methodological quality was assessed using the COSMIN Risk of Bias checklist. Measurement properties were evaluated according to COSMIN criteria, and the certainty of evidence was graded using a modified GRADE approach. Meta-analysis was performed when data were sufficiently comparable. Results: Seventeen studies met the inclusion criteria for narrative synthesis, of which eleven were included in a meta-analysis, representing fourteen cultural adaptations of the instrument. A descriptive random-effects meta-analysis of reported Cronbach’s α yielded a pooled estimate of 0.89 (95% confidence interval 0.85–0.92). This value should be interpreted cautiously, as structural validity was frequently insufficient across cultural adaptations, limiting interpretation of internal consistency according to COSMIN guidance. Other measurement properties, including reliability and measurement error, were frequently inadequately assessed or unreported. The certainty of evidence ranged from very low to moderate. Conclusions: Content validity was generally rated as sufficient, although certainty of evidence was low. Despite the high pooled α estimate, the reliability of Breast-CAM cannot be firmly established because structural validity was frequently insufficient across cultural adaptations. In accordance with the COSMIN ceiling rule, internal consistency was not considered sufficient in the absence of adequate structural validity. Key measurement properties, including test–retest reliability, measurement error, and responsiveness, were rarely evaluated. Further high-quality psychometric studies, particularly in culturally diverse populations, are needed to address these gaps and support appropriate use of the instrument in research and public health practice. Full article

This study presents a site-specific calibration of the ratio between S-wave and P-wave peak ground acceleration (PGA) for use in low-cost, on-site earthquake early warning (EEWS) systems in Iași, Romania. A dataset of 25 intermediate-depth Vrancea earthquakes (Mw 4.1–5.7; epicentral distances 150–210 km) was analyzed. PGA values were extracted for the P- and S-wave windows on both horizontal components and combined using geometric means. The resulting S/P amplitude ratios yield a median value of k_S/P = 6.19 and a logarithmic standard deviation of σlog₁₀ = 0.31, corresponding to a multiplicative uncertainty factor of approximately ×2. These results indicate that S-wave amplitudes are typically six times larger than P-wave amplitudes at this site, consistent with soft-soil amplification observed in comparable stations in Japan and Italy. The calibrated ratio can be used as a site-specific input for future MEMS-based on-site EEW implementations to estimate the expected S-wave PGA immediately after P-wave detection, with the observed S–P delays in Iași indicating a typical available warning window of 20–22 s. Full article

Diagnosis of cardiomyopathies often depends on overt phenotypic manifestations, delaying patient management. This underscores the need for population-level opportunistic screening tools using clinically indicated CT scans to detect subclinical myocardial disease. This study developed an Ensemble Machine Learning (ML) model to automatically segment the left ventricular myocardium from CT data and estimate the probability of underlying myocardial disease using radiomic feature analysis. A total of 60 CT scans (~12,000 images) were used to train ML models for left ventricular myocardium segmentation, including scans from both healthy individuals and patients with myocardial disease. A novel Ensemble model was developed and externally validated on 10 independent CT scans. Subsequently, 100 unseen CT scans were segmented manually and automatically for radiomic feature analysis. After removing highly correlated features through intra-class variation and correlation filtering, the refined dataset was used for model training and testing. Key predictive features were identified, and model performance was evaluated. The four best-performing models (Unet++, ED w/ASC, FPN, and TresUNET) were combined to form an Ensemble model, achieving a final DICE score of 0.882 after hyperparameter optimization. External validation yielded a DICE score of 0.907. Radiomic feature analysis identified 15 key predictors of myocardial disease in both manual and automatic segmentation datasets. The model demonstrated strong performance in detecting underlying myocardial disease, with AUCs of 0.85 and 0.8, respectively. This study presents a fully automated CT-based framework for LV myocardial segmentation and radiomic phenotyping that accurately estimates the probability of underlying myocardial disease. The model demonstrates strong generalizability across different CT protocols and highlights the potential role of AI-driven CT analysis for early, non-invasive cardiomyopathy screening at a population level. Full article