Search Results (1,651)

Background/Objectives: Dementia with Lewy bodies (DLB) is a clinically heterogeneous neurodegenerative disorder characterized by cognitive, neuropsychiatric, motor, and autonomic manifestations. Reduced striatal dopamine transporter availability on ¹²³I-ioflupane single-photon emission computed tomography (SPECT) is an established indicative biomarker of DLB. However, the associations among expert visual interpretation, regional semiquantitative dopamine transporter measurements, autonomic manifestations, and dementia severity remain incompletely characterized. This study aimed to evaluate the relationship between visual ¹²³I-ioflupane SPECT classification and regional DaTQUANT z-scores in patients with clinically probable DLB and to explore their associations with autonomic manifestations and dementia severity. Methods: This retrospective, cross-sectional study included 30 patients who received a clinical diagnosis of probable DLB according to the 2017 DLB Consortium criteria before the ¹²³I-ioflupane SPECT results became available. SPECT images were assessed visually by experienced nuclear medicine physicians and semiquantitatively using DaTQUANT software. Bilateral striatal, putaminal, caudate, and putamen-to-caudate ratio z-scores were analyzed in relation to visual scan classification, Clinical Dementia Rating (CDR) scores, and retrospectively ascertained autonomic manifestations, including orthostatic hypotension, delayed orthostatic hypotension, supine hypertension, postprandial hypotension, constipation, and urinary incontinence. Results: Scans were visually classified as supportive of nigrostriatal dopaminergic degeneration in 23 of 30 patients (76.7%) and as non-supportive in seven (23.3%). Bilateral striatal, putaminal, and caudate z-scores were significantly lower in visually supportive scans than in non-supportive scans (all p < 0.001), whereas putamen-to-caudate ratio z-scores did not differ significantly between the groups. None of the evaluated autonomic manifestations was significantly associated with either visual scan classification or regional DaTQUANT measurements. No regional DaTQUANT measurement was significantly associated with dementia severity. A modest positive correlation was observed between the left putamen-to-caudate ratio z-score and CDR (Spearman’s ρ = 0.373, nominal p = 0.050); however, given the small sample size and multiple regional comparisons, this borderline finding was considered exploratory. Conclusions: Regional DaTQUANT measurements were consistent with expert visual interpretation of ¹²³I-ioflupane SPECT in patients with clinically probable DLB. Associations of dopaminergic imaging measurements with autonomic manifestations and dementia severity were limited. Semiquantitative analysis may complement visual interpretation, but its findings should be interpreted within the broader clinical and biomarker context. Larger prospective studies incorporating standardized autonomic testing, appropriate control groups, longitudinal follow-up, and complementary biomarkers are warranted. Full article

This study focuses on the Significant Wave Height (SWH) retrieval from the Ka-band radar interferometer (KaRIn) on the Surface Water and Ocean Topography (SWOT) satellite by combining backscattering and interference characteristics. To this end, the backscattering-related and interference-related parameters were jointly used as inputs to develop a machine learning model. Here, the backscattering-related data include normalized radar cross-section (NRCS), incidence angle, and the image spectra parameters extracted from KaRIn Level 1B (L1B) data, while the interference-related data correspond to the Level 2 (L2) volumetric correlation, which characterizes the influence of ocean wave scattering on interferometric coherence. The machine learning model is built upon a Multi-Layer Perceptron (MLP), which serves as a nonlinear fitting tool. SWH retrievals from the proposed method and the existing L2 SWH product as a reference were validated by the collocated European Center for Medium-Range Weather Forecasts (ECMWF) reanalysis data, Haiyang2C (HY2C) and Haiyang2D (HY2D) altimeter data, and National Data Buoy Center (NDBC) buoy data. Validations show that both KaRIn SWH have a good agreement with collocations in terms of correlation coefficient (COR), BIAS and root mean square error (RMSE). Moreover, the retrieval accuracy from the proposed method (with an RMSE of about 0.29 m) is better than that of the L2 product (with an RMSE of about 0.46 m) when validated against the collocated ECMWF datasets. Ablation analysis further confirms that image spectra parameters and volumetric correlation are the dominant factors driving the retrieval accuracy improvement, with notable contribution differences among the sub-parameters of spectral features. This performance gain arises from the complementary physical mechanisms of backscattering and interferometric observables, which describe sea state information from independent dimensions. These accurate SWH retrievals can help correct sea state biases for collocated KaRIn sea surface height products and complement wave products from other satellite sensors. Full article

Underwater visual degradation makes it difficult for the image modality to represent video semantics, while label sparsity in underwater scenes leads to weak inter-category correlations, thereby degrading the performance of multi-label classification. To address these issues, this paper proposes a Semantic-Enhanced Underwater Videos Multi-Label Classification Network Based on Structural Graph Convolution (SEMGCN). Specifically, the proposed method first disentangles the image and text modalities into shared and private representations, and enhances feature representation capability through orthogonal constraints and feature reconstruction. Moreover, a Cross-Modal Category-Aware Module (CCAM) is constructed to model interactions between image and text features and perform bidirectional cross-attention with category-label text embeddings, thereby generating category-aware initial node representations. Furthermore, to alleviate the limitation of semantic propagation caused by sparse label co-occurrence, a Structural Graph Convolutional Network (SGCN) is proposed. By integrating explicit co-occurrence relationships with implicit structural similarity relationships, the proposed model collaboratively captures both explicit and latent semantic associations, thereby improving multi-label classification performance under label-sparse conditions. Experiments were conducted on the self-constructed Underwater Video Multi-label Classification Dataset (UVMC) and the public MLSV2018 dataset. The experimental results show that SEMGCN achieves Average Precision scores of 0.8645 and 0.8388 on UVMC and MLSV2018, respectively, demonstrating the effectiveness of the proposed method. Full article

Background/Objectives: The magnetic resonance imaging (MRI)-derived scapular superomedial angle (SMA) may serve as a continuous descriptor of scapular geometry. We evaluated whether SMA is associated with acromial morphology, examined SMA and subacromial space (SAS) across acromial categories, and quantified measurement reliability. Methods: In this retrospective cross-sectional study, 306 shoulders from 279 patients (27 bilateral) were analyzed after screening 353 examinations. SMA and SAS were measured independently by two orthopedic observers, and acromial morphology was classified independently by two radiologists, all in blinded sessions. Reliability, group comparisons, receiver operating characteristic (ROC) analyses, and patient-level clustered generalized estimating equations were performed. Results: Interobserver reliability was excellent for SMA (intraclass correlation coefficient [ICC], 0.960) and SAS (ICC, 0.926), and intraobserver reliability was excellent for SMA and good for SAS (ICC, 0.944 and 0.848); acromial classification agreement was almost perfect (linear-weighted kappa, 0.980). SMA increased stepwise across Type I to III acromia (128.7°, 133.9°, 137.2°; p < 0.001), whereas SAS decreased (7.5, 6.8, 6.3 mm; p < 0.001). Each 1° SMA increase was independently associated with Type III morphology (adjusted odds ratio, 1.18; p < 0.001). SMA was not independently associated with rotator-cuff severity (p = 0.842). Because adjacent Type II–III differences approached the minimum detectable change, individual-level discrimination was limited. Conclusions: MRI-derived SMA is a reproducible descriptor associated with acromial morphology at the group level, but adjacent-category differences approached measurement error. SMA should not substitute for direct acromial classification, and external validation is required before clinical thresholds are considered. Full article

Background/Objectives: Accurate preoperative biometry is essential for refractive cataract surgery. Pentacam AXL WAVE is an emerging multimodal platform integrating ocular biometry, anterior segment imaging, and corneal assessment, yet its interchangeability with established devices remains unclear. This study aimed to evaluate the agreement and clinical interchangeability of Pentacam AXL WAVE with IOLMaster 700 and Pentacam. Methods: This retrospective cross-sectional study compared Pentacam AXL WAVE with IOLMaster 700 and Pentacam across 175 eyes of 105 cataract patients. Interdevice agreement was evaluated using intraclass correlation coefficients, concordance correlation coefficients, and Bland–Altman analysis with multiple outlier-handling strategies; generalized estimating equations were applied to identify clinical factors associated with interdevice differences. Results: Pentacam AXL WAVE demonstrated good agreement with IOLMaster 700 and Pentacam for axial length and anterior corneal curvature parameters, but showed weaker agreement for white-to-white diameter, total keratometry, total corneal astigmatic vectors, and several extended anterior segment parameters. Fewer clinical factors explained interdevice differences versus IOLMaster 700, whereas ocular surface or corneal abnormalities, refractive status, and cataract characteristics were associated with greater differences versus Pentacam. Conclusions: These findings indicate that Pentacam AXL WAVE has parameter-specific interchangeability boundaries and can meet some routine biometric needs; however, for eyes requiring high refractive precision or presenting with complex ocular conditions, interpretation should be based on the agreement characteristics of individual parameters. Full article

Background/Objectives: Deep learning-based medical imaging applications often require large and diverse datasets to achieve reliable performance. However, publicly available palpebral conjunctiva datasets for non-invasive hemoglobin assessment remain limited in both size and demographic diversity. The objective of this study was to develop and evaluate a conditioning-guided Latent Diffusion Model (LDM) for controllable palpebral conjunctiva image synthesis under limited-data conditions. Methods: The proposed framework employs an image-to-image latent diffusion strategy conditioned on continuous hemoglobin values together with gender and country information to generate realistic synthetic conjunctiva images with controllable clinical and demographic characteristics. The proposed LDM was compared with GAN-based approaches, including cDCGAN and StyleGAN2-ADA, using evaluation criteria covering image realism, diversity, conditioning consistency, and computational efficiency. Analyses of frequency-domain characteristics, zero-shot cross-population evaluation, and blinded clinical assessment were performed for the proposed LDM. Results: The proposed LDM achieved the lowest FID score ($15.86\pm 0.23$) among the evaluated models, indicating superior image realism relative to cDCGAN and StyleGAN2-ADA, while maintaining image diversity comparable to StyleGAN2-ADA and substantially outperforming cDCGAN. Conditioning evaluation demonstrated strong consistency between the target hemoglobin values and the generated images, achieving a Pearson correlation coefficient of $r=0.910\pm 0.025$. Frequency-domain analysis, zero-shot cross-population evaluation, and blinded clinical assessment further supported the realism, structural consistency, and clinical plausibility of the generated images. Conclusions: The findings demonstrate the potential of conditional latent diffusion models for controllable palpebral conjunctiva image synthesis under limited-data conditions. The proposed framework provides a promising approach for generating realistic synthetic conjunctiva images with controllable clinical and demographic characteristics. Full article

Text-based person search (TPS) suffers from cross-modal informational skewness: pedestrian images are high-dimensional and redundancy-prone, while textual descriptions are sparse, incomplete, and sometimes inaccurate. To address the low alignment accuracy and poor robustness caused by the inherent uneven information distribution of visual and textual modalities in TPS, this paper proposes a unified Partition-based Information Rebalancing (PaIR) framework to realize balanced optimization and precise alignment of cross-modal information from both global content and local part dimensions. The framework adopts the CLIP dual-modal encoder for basic feature extraction and constructs a parallel global–local dual representation system to compensate for the lack of fine-grained spatial information in single global features. To eliminate modal redundancy and noise interference, a dual-modal noise suppression module is designed to filter invalid redundant information through visual foreground–background separation and textual token weight screening, while introducing adversarial constraints and orthogonal constraints to purify effective features. On this basis, a part balance alignment module is built to complete human semantic part decomposition and soft matching alignment for dual-modal features. Aiming at the common part semantic missing problem in textual descriptions, a visual part correlation affinity matrix is utilized for semantic associative completion to balance the information density of dual modalities. Finally, a global–local joint alignment strategy integrates hierarchical features and bidirectional cross-modal attention interaction to eliminate global–local semantic discontinuity and enhance fine-grained cross-modal matching capability. Extensive experiments on three public benchmarks demonstrate that PaIR consistently improves multiple baselines. Full article

Objective: The aim of the study was to sonographically evaluate the thickness of the abdominal, multifidus lumborum and gluteal muscles in professional dancers and the relationship between anthropometric, demographic and professional characteristics and clinical tests of the hip and groin. Methods: A cross-sectional study was employed, examining ultrasound imaging in 20 elite professional ballet dancers to assess the thickness of the abdominal muscles, lumbar multifidus, gluteus medius and gluteus minimus. The associations between the ultrasound variables and hip and groin function were tested using the Copenhagen Hip and Groin Outcome Score (HAGOS) and the Beighton score. The Spearman correlation coefficient was used to find the relationship between muscle thickness, Beighton score and HAGOS. Results: The increase in muscle thickness from rest to activity was greater in the transversus abdominis left (62.6%) and the internal oblique left (38.7%). The strongest correlations between muscle thickness and HAGOS were observed for the gluteus medius left at rest (0.55) and active (0.58) in Physical Function in Sport and Recreation and right active in Symptoms (0.42), Physical Function in Sport and Recreation (0.48), and Participation in Physical Activities (0.54). The correlation of the Beighton score with muscle thickness was inverse for the internal oblique left at rest (−0.50) and active (−0.48), the transversus abdominis active left (−0.52) and right (−0.56), the rectus abdominis right at rest (−0.55), and the gluteus minimus right at rest (−0.51). Conclusions: The muscle thickness of the gluteus medius was positively correlated with the HAGOS questionnaire, whereas joint hypermobility was negatively correlated with internal oblique, transversus abdominis, rectus abdominis and gluteus minimus muscle thickness. Ultrasound imaging should be considered for the measurement of muscle thickness in hip and groin pain. Trial Registration: ClinicalTrials.gov, NCT0565416. Date of registration: 8 December 2022. Full article

Background: Interstitial fibrosis is a hallmark of chronic kidney disease (CKD) progression; however, its assessment currently relies on kidney biopsy, which is invasive and prone to sampling error. Fibroblast activation protein inhibitor (FAPI) positron emission tomography/computed tomography (PET/CT) targets activated fibroblasts and may enable non-invasive assessment of renal fibrosis. We investigated whether renal parenchymal [⁶⁸Ga]Ga-FAPI uptake reflects histological fibrosis and evaluated its performance in highly fibrotic kidneys. Methods: In this prospective single-center study, 23 patients undergoing clinically indicated kidney biopsy underwent [⁶⁸Ga]Ga-FAPI PET/CT imaging. Standardized uptake values (SUV) of the kidneys were matched to histological fibrosis assessments as the H-score (0–300), the fibrotic area (%) and intensity grade (0–III). Associations were calculated by Spearman rank correlation and diagnostic performance for advanced fibrosis was evaluated by ROC analysis with leave-one-out cross-validation. Results: SUVmean correlated with H-score (ρ = 0.71), fibrotic area (ρ = 0.79) and intensity grade (ρ = 0.74; all p < 0.001), whereas SUVmax (ρ = 0.17) and SUVpeak (ρ = 0.29) did not. For advanced fibrosis (intensity grade III), SUVmean discriminated with an AUC of 0.89 (95% CI 0.67–1.00). A Youden-derived cut-off of ≥2.83 yielded cross-validated sensitivity/specificity of 92%/82%. Conclusions: Renal parenchymal FAPI expression, represented by SUVmean, correlates with the histological severity of interstitial fibrosis, supporting [⁶⁸Ga]Ga-FAPI PET/CT as a non-invasive tool to potentially monitor renal fibrosis over time. FAPI expression is closely linked to renal function, and subsequent larger studies are needed to validate these findings and the proposed exploratory cut-off value. Full article

Accurate cloud segmentation from ground-based fisheye camera imagery is essential for solar irradiance forecasting and photovoltaic system optimization. Traditional computer vision approaches, such as HSV thresholding and K-means clustering, face significant limitations when applied globally to sky images due to the spectral similarity between cloud regions and sky areas under varying atmospheric conditions. This study presents a hybrid methodology that leverages YOLOv8 instance segmentation to provide contextual cloud regions followed by refined HSV thresholding within these detected areas. The approach incorporates solar trajectory modeling using pvlib for accurate sun disk detection and exclusion, preventing false cloud classification. The methodology was developed and validated at the CIESOL using Mobotix Q71 fisheye cameras, and later tested in Antofagasta (Chile) and Huelva (Spain). The YOLOv8l-seg model achieved a mask precision of 0.821 and box mAP@0.5 of 0.680 on validation data. The results show a promising correlation with radiometric measurements such as clearness index $k_t$ and diffuse fraction $k_d$ in preliminary validation cases. While YOLOv8 demonstrates good cross-site generalization, HSV thresholding requires camera-specific calibration for optimal performance. The method addresses the context-dependency limitations of traditional algorithms, though computational performance and broader validation remain areas for future work. Full article

Background: Metabolic dysfunction-associated steatotic liver disease (MASLD) is associated with ectopic fat accumulation and alterations in adipose tissue function. However, the relationships of structural and thermogenic adipose imaging markers with hepatic steatosis remain incompletely understood. This study aimed to jointly evaluate the cross-sectional associations of epicardial adipose tissue (EAT) thickness and infrared thermography (IRT)-derived brown adipose tissue (BAT)-related thermogenic activity with controlled attenuation parameter (CAP)-defined hepatic steatosis severity in adults with suspected MASLD. Methods: In this cross-sectional study, 207 adults undergoing clinical evaluation for suspected MASLD underwent transient elastography to obtain the controlled attenuation parameter (CAP) for hepatic steatosis assessment. EAT thickness was measured by transthoracic echocardiography, and BAT-related thermogenic activity was assessed by infrared thermography using the supraclavicular-to-chest temperature difference (ΔTemp). Associations of these adipose imaging phenotypes with CAP were evaluated using correlation analyses, sequential multivariable linear regression models, and BAT-stratified analyses. Results: EAT thickness increased progressively across CAP-defined steatosis grades (p < 0.001) and was positively correlated with CAP (r = 0.637, p < 0.001), whereas ΔTemp decreased with increasing steatosis severity and was inversely correlated with CAP (ρ = −0.277, p < 0.001). In sequential multivariable regression models, EAT thickness remained independently associated with CAP across adjustments for age, sex, body mass index, metabolic variables, and ΔTemp (standardized β = 0.559–0.623; all p < 0.001). Both EAT thickness and ΔTemp were independently associated with CAP in the fully adjusted model, with a stronger association for EAT thickness (standardized β = 0.559 vs. −0.167; p < 0.001 and p = 0.004, respectively). Stratified analyses demonstrated consistent associations between EAT thickness and CAP across both BAT-low and BAT-high activity groups. Conclusions: Greater EAT thickness and lower IRT-derived ΔTemp were independently associated with greater CAP-defined hepatic steatosis severity, with EAT thickness showing the stronger standardized association. These complementary structural and thermogenic imaging correlates warrant prospective evaluation to clarify their directionality and clinical relevance. Full article

Reliable crop-type mapping from satellite image time series is affected by distribution shifts across geographic regions, agricultural years, and heterogeneous label systems. To address this challenge, we propose Cross-Domain Mixup (CDMix), a supervised domain-adaptation method designed to leverage a larger labeled source dataset to improve performance on a smaller labeled target dataset under distribution shifts. We also introduce PixelSet-Slovakia, a new multi-year, parcel-level Sentinel-2 dataset covering three Slovak study regions and several growing seasons. Using a common backbone, we compare CDMix against three families of adaptation strategies: (i) no adaptation, (ii) weight transfer through fine-tuning and encoder freezing, and (iii) feature-space alignment using Maximum Mean Discrepancy (MMD) and Correlation Alignment (CORAL). All methods are evaluated in two scenarios: geographic supervised adaptation across datasets from two countries and temporal supervised adaptation across different growing seasons. Across both tested source–target settings, CDMix generally achieves competitive performance when initialized from pretrained representations. Under the region-held-out validation protocol, several pretrained adaptation strategies outperform training from scratch. Full article

Hyperspectral image super-resolution (HSI SR) focuses on enhancing the spatial resolution of HSIs while preserving their inherent spectral information. Existing single-image HSI SR methods still suffer from blurred spatial edges and spectral distortion. Although numerous spatial–spectral enhancement networks can enhance spatial–spectral feature extraction, they often lead to a cumbersome network architecture. To address these issues, we propose a Spatial–Spectral Distillation Mamba Network, called SSDM-Net, for HSI SR, which contains a main reconstruction branch and two training-only auxiliary branches for spatial and spectral knowledge distillation. Specifically, the spatial and spectral auxiliary branches, which are utilized exclusively during training, provide edge-aware guidance and capture spectral correlations, respectively. During training, the spatial–spectral knowledge is transferred to the main branch. During inference, the auxiliary branches are removed, improving reconstruction quality without extra computational burden. In the main branch, a Mamba-based spatial–spectral global enhancement module processes spatial and latent inter-channel sequences using selective scanning whose cost is linear in the processed sequence lengths when the feature dimensions are fixed. In addition, a dynamic loss weighting strategy is developed to balance reconstruction, distillation, and auxiliary losses during optimization. Comprehensive experiments conducted on the CAVE and Houston datasets with three scale factors demonstrate that SSDM-Net produces more accurate reconstruction results than existing representative HSI SR methods. Cross-dataset experiments on the Harvard dataset further suggest that the method can maintain competitive reconstruction performance under the evaluated cross-dataset settings. Full article

Repeated non-equidistant unmanned aerial vehicle (UAV) inspections of slope-road cracks require measurements from different distances, poses, and image scales to remain comparable and sufficiently precise for engineering-state decisions. Existing studies rarely integrate cross-view physical conversion, measurement uncertainty, and a project-defined minimum detectable change. We develop an engineering-detectability framework that defines cross-period criteria for crack width and displacement and derives equivalent widths for ideal, representative non-standard, and arbitrary viewpoints. First-order error propagation and reliability allocation convert the minimum detectable change into accuracy requirements for range, field of view, and normalized image coordinates. Crack-boundary coordinates and localization uncertainties provide a common interface for interchangeable detection and photogrammetric modules. Validation combines a controlled fixed-camera sequence with a close-range field-camera multiview test of seven physical openings under local coplanarity. All six determinate stages in the controlled sequence agreed with the digital image correlation (DIC) comparison, while one borderline stage required review. Across the seven openings, the four-view means gave a mean absolute error (MAE) of 0.196 mm and a root mean square error (RMSE) of 0.270 mm, with cross-view coefficients of variation (CVs) of 0.33–4.93%. An illustrative error budget demonstrates reverse screening of system configurations from project thresholds. The framework therefore connects viewpoint-equivalent measurements, uncertainty constraints, and engineering-state decisions in an auditable chain. Full article

Sustainable historic-town conservation requires preserving historical patterns and cultural memory while understanding how spatial environments shape everyday perception. Using Yun’an Ancient Salt Town in Chongqing, China, this study develops a three-layer framework of spatial topology, scene interfaces, and historical semantic elements. Space syntax, deep-learning image recognition, rule-based coding, image-based questionnaires, Spearman correlation, random forest regression, and SHAP were used to examine perceived historicity, safety, attractiveness, and comfort. The analysis combined axial models for 1985, 2004, and 2024 with 140 images—55 human view and 85 aerial view—each evaluated by 264 valid respondents. Yun’an’s street network shifted from a salt-production and transport structure toward modern traffic corridors; integration declined in the historic core, and intelligibility fell from 0.13 in 2004 to 0.07 in 2024. At the human-view scale, modern interference was negatively associated with historicity and attractiveness, whereas stairs, traditional components, and micro-historical objects were positively associated with historicity. At the aerial-view scale, historical visibility, character integrity, and blue-green-space indicators were associated with historicity, attractiveness, and comfort. Cross-validated random forest performance was strongest for aerial-view historicity and limited for other outcomes. The findings support a human-centered, culturally sustainable renewal pathway linking historical-street continuity, townscape integration, blue-green quality, and heritage-node activation. Full article