Search Results (5,851)

Worldwide visual geo-localization aims to predict the geographic coordinates of an image capture location from visual content alone, posing unique challenges due to the vast scale of the Earth’s surface and pervasive visual ambiguity across distant regions. Existing approaches face distinct limitations as follows: retrieval-based methods demand massive geo-tagged databases and scale poorly; alignment-based models lack interpretability and are vulnerable to visually similar scenes; and large vision-language models (LVLMs) offer semantic reasoning but suffer from hallucination. A natural solution is retrieval-augmented generation (RAG), yet we observe that directly injecting retrieved candidates as context causes severe context poisoning. To address this, we propose HybridGeo, a dual-stream late-fusion framework that decouples retrieval from reasoning. A retrieval stream applies continuous alignment with spatial–semantic clustering to produce stable regional anchors; a reasoning stream performs context-free Chain-of-Thought inference to yield an independent coordinate estimate. The two streams are fused only at the decision stage via a spatial–consistency module that triggers weighted averaging under agreement or confidence-based arbitration under conflict. Experiments on Im2GPS3k show that HybridGeo achieves 73.89% Country@750km accuracy, outperforming the retrieval baseline by 7.27% and 8.23%, and surpassing both VLM-only and RAG baselines. These results demonstrate that late fusion effectively avoids context poisoning while enabling complementary benefits from both streams. Full article

Background: Left ventricular global longitudinal strain (GLS) measured by speckle-tracking echocardiography (STE) has become a key marker of myocardial systolic function, yet normal reference values remain heterogeneous, and the magnitude of physiological sex differences is not fully defined. We performed a systematic review and meta-analysis to establish pooled GLS reference estimates in healthy individuals, quantify sex-related differences, and contextualize deformation findings relative to conventional systolic function. Methods: A systematic search of PubMed, Scopus, and EMBASE identified observational studies reporting GLS in healthy adults assessed by two-dimensional or three-dimensional STE. Random-effects meta-analysis using standardized mean differences (SMD) compared GLS between women and men. Descriptive pooled reference values were derived using weighted median and interquartile range (IQR) reconstruction from study-level distributions. Meta-regression analyses explored demographic, clinical, and methodological sources of heterogeneity. A complementary analysis evaluated sex-related differences in left ventricular ejection fraction (LVEF) within the same populations. Results: Thirty-two studies, including 19,157 healthy individuals, were analyzed. The pooled population had a weighted median age of 47.5 years and 53% female participants. Overall, GLS demonstrated a weighted median of 20.3% (IQR 17.8–22.5). Women showed higher GLS values than men (20.8% [18.4–23.1] vs. 19.4% [17.0–21.6]). Meta-analysis of 28 studies confirmed significantly greater GLS in females (SMD 0.487, 95% CI 0.409–0.565; p < 0.001), with consistent findings across imaging modalities and no subgroup interaction. Between-study heterogeneity was substantial (I² = 82.7%), although effect direction was uniform. Meta-regression analyses identified no significant moderators, and sensitivity analyses confirmed stable estimates without publication bias. Segmental analysis demonstrated a physiological base-to-apex strain gradient. In contrast, LVEF was largely comparable between sexes, with no clinically meaningful difference (SMD 0.257, 95% CI 0.186–0.327; p < 0.001), indicating preserved global systolic performance despite differences in myocardial deformation. Conclusions: GLS demonstrates a consistent physiological range in healthy populations, with women exhibiting higher longitudinal deformation than men, independent of the imaging modality. These findings support the adoption of sex-specific GLS reference values and highlight the complementary roles of deformation and volumetric indices in improving the interpretation of myocardial function and reducing misclassification in clinical practice. Full article

The rapid proliferation of text-to-image generative systems has transformed visual content production, yet the structural characteristics embedded in their compositional outputs remain insufficiently understood. Rather than approaching human–AI differentiation as a purely classification problem, this study investigates whether a controlled set of AI-generated and human-designed posters exhibits measurable structural divergence under thematically matched conditions. A dataset of jazz festival posters was analyzed using interpretable geometric and information-theoretic descriptors, including spatial density (padding ratio), edge density, chromatic dispersion, and entropy-based measures. Instead of relying on deep neural detection architectures, we employed a transparent machine-learning framework to examine intrinsic structural separability within feature space. Results demonstrated highly stable group separation (ROC-AUC = 0.99; 95% CI: 0.978–0.999) under cross-validated evaluation. Distributional analysis further revealed a pronounced divergence in spatial density allocation (Kolmogorov–Smirnov statistic = 0.76, p < 10⁻²⁸), accompanied by a very large effect size (Cohen’s d = 1.365). While padding ratio emerged as the dominant discriminative factor, additional entropy- and chromatic-based descriptors contributed to group separation even when spatial density was excluded (AUC = 0.903). These findings indicate that AI-generated and human-designed posters can diverge in negative space allocation and chromatic organization under controlled thematic and platform-specific conditions. The study contributes to the explainable analysis of generative visual systems by reframing human–AI differentiation as a structural divergence problem grounded in interpretable image statistics rather than as a model-specific artifact detection task. Full article

Artificial intelligence (AI) is increasingly used to support tuberculosis (TB) screening and diagnosis, particularly through computer-aided detection (CAD) applied to chest radiography (CXR). However, the programmatic value of AI depends not only on diagnostic accuracy but also on implementation context, threshold calibration, and integration into diagnostic pathways. We conducted a narrative, state-of-the-art review of AI applications across the TB diagnosis pathway. Evidence was synthesized from World Health Organization policy documents, independent validation initiatives, and peer-reviewed studies published between 2010 and 2026, with a structured selection process aligned with PRISMA principles. CAD for CXR is the most mature AI application and is recommended by WHO for TB screening and triage among individuals aged ≥15 years in specific contexts. Across studies, CAD-CXR demonstrates sensitivity comparable to human readers, although performance varies by product, population, and imaging conditions, necessitating local threshold calibration. Evidence from implementation studies suggests improvements in screening efficiency and potential cost-effectiveness in high-burden settings. Other AI modalities, including computed tomography (CT)-based imaging analysis, point-of-care ultrasound interpretation, cough or stethoscope sound analysis, clinical risk models, and genomic resistance prediction show promising but heterogeneous results, with most requiring further independent validation and prospective evaluation. AI has the potential to strengthen TB screening and diagnostic pathways, but its impact depends on integration into health systems and evaluated using patient- and program-level outcomes rather than accuracy alone. A differentiated approach is needed, with responsible scale-up of policy-endorsed tools alongside rigorous evaluation of emerging technologies to support effective and equitable TB care. Full article

In the context of urban renewal, the spatial interface of historic commercial districts functions as both a carrier of historical character and a key setting for commercial activity, public life, and local cultural expression. To address the limitations of conventional studies that rely heavily on field observation and qualitative description, this study takes Xiaohe Zhijie in Hangzhou as a case and develops a multi-level street-view recognition framework for the quantitative analysis of spatial interface characteristics. Based on street-view image collection and standardized preprocessing, a sample database was established at the sampling-point scale. Semantic segmentation, automated commercial object detection, and manual interpretation were combined to identify interface elements, including buildings, sky, greenery, pavement, vehicles, pedestrians, and commercial objects, while commercial content was assessed in terms of locality and homogenization. The results show that Xiaohe Zhijie exhibits a building-dominated and relatively enclosed interface pattern, with greenery and pavement forming the basic environmental ground, weak vehicle interference, and localized enhancement of vitality through commercial objects and pedestrian activities. Significant differences were found among street segments in openness, commercial coverage, and local expression. Three interface types were identified: commercial–cultural composite, local life-oriented, and waterfront landscape–cultural composite. The main challenge lies not in commercialization itself, but in stronger visual locality than content locality and increasing homogenization, resulting in a pattern of “localized form but homogenized content.” Full article

Median overall survival remains a central endpoint in oncology, but it can obscure a clinically meaningful long tail of patients with advanced solid tumors who survive well beyond the median. One biological context in which this pattern may be relevant is tumor microenvironment (TME) acidosis. Driven by aerobic glycolysis, hypoxia, impaired perfusion, and proton-export programs, acidic TME is increasingly implicated in invasion, therapeutic resistance, and immune suppression. This narrative review examines TME acidosis as the primary biological framework and considers long-tail survival as a clinical lens through which its implications may be interpreted. We summarize the biological basis and heterogeneity of acidic TME, review current approaches to clinical and translational assessment of tumor acidity, including acidoCEST magnetic resonance imaging (MRI) and positron emission tomography (PET)-based approaches, and discuss the potential and limitations of alkalization-oriented interventions such as buffering and diet-based strategies. Particular attention is given to the distinction between direct measurements of tumor acidity and clinically feasible but indirect markers such as urinary pH, which should not be interpreted as a direct surrogate for local tumor extracellular pH. From a science-based medicine perspective, long-tail survival is treated here as a hypothesis-generating clinical signal rather than proof of causality. Overall, alkalization-oriented interventions appear biologically plausible and clinically testable, but current clinical evidence remains limited and context-dependent. Future progress will require mechanistically informed biomarkers, careful safety evaluation, and trial designs capable of detecting delayed separation of survival curves and tail-oriented patterns of benefit. Full article

Background: Low-dose computed tomography (LDCT) lung cancer screening (LCS) frequently identifies emphysema in high-risk smokers. However, the extent to which CT-detected emphysema reflects underlying physiological impairment remains uncertain. We evaluated whether spirometry can detect functional abnormalities in this population beyond structural imaging findings. Methods: This cross-sectional study included 323 individuals with LDCT- detected emphysema and no lung cancer or prior chronic respiratory diseases within a screening cohort (n = 3076). Participants underwent pre-bronchodilator spirometry and symptom assessments (COPD Assessment test (CAT) and Modified Medical Research Council (mMRC) Dyspnea Scale). Pre-bronchodilator airflow limitation was defined as forced expiratory volume in one second to forced vital capacity ratio (FEV₁/FVC) < 0.70. Small airways dysfunction was defined by ≥2 reduced mid-expiratory flow parameters (<60% predicted). Flow–volume curve morphology was assessed qualitatively. Results: Pre-bronchodilator airflow limitation was observed in 45.2% of participants, predominantly mild. Small-airway dysfunction was present in 52%, and an abnormal flow–volume curve morphology in 67.5%. Notably, functional abnormalities were frequently observed despite preserved FEV₁. Symptom burden was low, with only 7.7% of participants reporting clinically significant symptoms. Functional impairments often overlapped and were common in minimally symptomatic individuals. Conclusions: In a lung cancer screening (LCS) cohort with CT-detected emphysema, functional abnormalities are frequently observed, including in individuals with preserved FEV₁ and minimal symptoms. Spirometry provides additional physiological insight beyond structural imaging; however, these findings are descriptive and should not be interpreted as diagnostic of COPD. Further studies are needed to determine their clinical relevance. Full article

Glioblastoma (GBM) remains among the most treatment-refractory human malignancies. It is characterized by profound radioresistance and a highly immunosuppressive tumor microenvironment, limiting the durable efficacy of radiotherapy. Beyond direct cytotoxicity, ionizing radiation can induce immunogenic cell death and the release of damage-associated molecular patterns (DAMPs), including surface-exposed calreticulin, HMGB1, extracellular ATP/adenosine, and tumor-derived DNA. These signals engage pattern-recognition receptors and cGAS–STING–type I interferon pathways, transiently promoting antigen presentation and immune activation. In GBM, however, DAMP signaling frequently evolves toward chronic inflammation and immune suppression, characterized by myeloid cell recruitment, adenosine accumulation, and immune checkpoint upregulation, thereby contributing to tumor regrowth and radioresistance. This dual immune-regulatory role of DAMPs highlights the importance of temporal and contextual interpretation of radiation-induced immune responses. In this review, we summarize current mechanistic and translational evidence on DAMP-mediated immunomodulation in GBM radiotherapy; discuss modality-dependent considerations across photon, proton, and high-LET irradiation; and evaluate the emerging potential of DAMPs as dynamic biomarkers of treatment response. We further outline how integration of DAMP profiling with liquid biopsy, imaging, and nanotheranostic platforms may support biologically informed and adaptive radiotherapy strategies for glioblastoma. Full article

This paper examines the iconography of the Mother of God holding a book in Early Christian and Medieval art, focusing on representations in which a book or scroll functions as an attribute of the Virgin Mary. Particular attention is given to scenes depicting Mary in relation to the Christ Child, Christ Pantocrator, and the Magnificat. The study explores the symbolic significance of the book and scroll through the textual tradition of the Church Fathers. Adopting the methodological approach to the iconographical structure developed by André Grabar, the paper centers on three interconnected case studies. First, it offers a close re-examination of a Marian scene on the ivory relief of the Werden casket (9th c.) of which the meaning is hard to understand. Second, it analyzes the depiction of the Mother of God in the vault of the crypt of Epiphanius at San Vincenzo al Volturno (9th c.), with particular emphasis on motifs that associate the image with the theme of Mary’s Triumph. Finally, it considers a fresco of Mary and Christ enthroned from the Egyptian monastery of Deir al-Suryan (10th c.), treating these works as semantically and conceptually related. Through this comparative analysis, the paper advances several interpretations of the Magnificat as articulated in Early Christian visual culture and developed in later periods with the contribution of the Byzantine theology. Given the well-established influence of Early Christian art on both the Carolingian Renaissance in the West and the Byzantine East, the shared iconographical details identified here—both formal and conceptual—are understood as deriving from a common visual tradition rooted in Antiquity. Full article

Background: Glioblastoma remains the most aggressive and treatment-resistant primary brain tumor, with patient outcomes strongly associated with the extent of surgical resection. Tumor recurrence is largely driven by infiltrating glioma cells that extend beyond the contrast-enhancing margin, which has traditionally served as the boundary for surgical resection. Advances in pre- and intraoperative imaging, functional mapping, and fluorescence guidance have challenged the conventional definition of “maximal safe resection” and given rise to the concept of supramaximal resection (SMR). This technique, where surgical resection extends beyond the contrast-enhancing border, has garnered significant interest in recent years and shown promising preliminary survival outcomes. However, the lack of standardized definitions and methodological consistency has limited reproducibility and clinical adoption. Methods: A systematic literature search of PubMed/MEDLINE, Embase, and Web of Science was performed from database inception through March 2026 in accordance with PRISMA guidelines. Studies investigating resection beyond the contrast-enhancing tumor margin in adult glioblastoma patients were evaluated for inclusion. Results: A total of 1045 records were identified, with 37 studies meeting inclusion criteria. Across studies, SMR was frequently associated with improved progression-free and overall survival in selected patients, particularly following complete contrast-enhancing tumor resection. However, substantial heterogeneity exists in SMR definitions, and the current body of evidence is largely retrospective and derived from high-volume centers. Conclusions: SMR represents a promising extension of maximal safe resection targeting infiltrative tumor beyond conventional imaging boundaries. While emerging evidence suggests survival benefits, variability in methodology and patient-specific factors require cautious interpretation. Future standardization and prospective validation are needed to better define the role of SMR within multimodal glioblastoma treatment. Full article

Objective: To develop and validate an artificial intelligence (AI) system for automated lens opacities classification system III (LOCS III)-based grading of all three major cataract subtypes using anterior segment optical coherence tomography (AS-OCT). Methods: This is a single-center cross-sectional study. AS-OCT images were collected and manually graded by ophthalmologists according to LOCS III. The dataset was randomly split into training, validation, and test sets. We propose a novel multi-granularity mask-guided network (MMNet) that jointly performs lens substructure segmentation and severity grading. The model’s performance was assessed on an independent test set for automatic grading of cortical cataract (CC), nuclear cataract (NC), and posterior subcapsular cataract (PSC) and the grading performance of the proposed method against ophthalmologists was also evaluated. The model’s interpretability was assessed via attention heatmaps and feature visualization. Results: The proposed MMNet exhibited high agreement with ground truth conducted through gold standard. The proportions of predictions with an absolute error < 1.0 for three subtypes range from 83.02% to 89.94%. The model’s grading accuracy for cataract subtypes was between 82.20 ± 1.41% and 89.76 ± 1.31% among the three subtypes, the Area Under the Curve (AUC) was between 0.954 (95% CI, 0.952–0.969; p < 0.001) and 0.973 (95% CI, 0.964–0.985; p < 0.001). The MMNet shows a satisfactory mean absolute error (MAE) of 0.14 ± 0.35 in CC, 0.10 ± 0.30 in NC, and 0.17 ± 0.38 in PSC grading. It also achieved a fast grading speed of 0.0178 s/image against manual grading. Conclusions: The proposed AI model presented advanced performance on AS-OCT images in automated LOCS III-based cataract grading for CC and NC, and also showed feasibility in PSC assessment. Full article

Gliomas are generally readily detected and broadly characterized using conventional MRI; however, substantial challenges remain in accurately delineating tumor extent, grading heterogeneous disease, and translating imaging findings into consistent, reproducible clinical decisions. Despite reported Dice coefficients of 0.85–0.91 for whole-tumor segmentation and classification AUC values exceeding 0.90 for glioma grading in curated datasets, most AI systems remain limited by validation design, dataset bias, and inadequate external generalizability. This narrative review synthesizes current AI applications for MRI-based glioma detection and segmentation, highlighting the evolution from radiomics-based classical machine learning approaches relying on handcrafted features to deep learning models capable of end-to-end representation learning. Commonly used MRI sequences, algorithmic paradigms, and reported performance trends are reviewed, with particular emphasis on tumor segmentation as a foundational enabling task. Key limitations that hinder clinical translation are examined, including limited dataset diversity, validation practices that inflate reported performance, domain shift across institutions, acquisition-related bias, and inadequate model interpretability. Emerging strategies to address these challenges, such as multi-institutional training, harmonization techniques, explainable AI frameworks, and workflow-integrated validation, are also discussed. While AI-based models demonstrate strong technical performance in research settings, their clinical impact will depend on rigorous external validation, transparency, and alignment with real-world neuro-oncology workflows. Full article

As AI-driven educational systems increasingly rely on deep neural networks, their vulnerability to adversarial perturbations raises concerns about assessment integrity, fairness, and reliability. Adversarial example purification is attractive for such deployments because it removes input perturbations without modifying the already deployed models. However, most existing purification methods are inherently goal-free: denoising-based approaches apply blind heuristic operators, while reconstruction-based methods rely on stochastic sampling guided by natural image priors. These methods typically suppress perturbations at the cost of weakening semantic details or inducing structural distortions. To address this limitation, we propose a novel goal-directed purification framework, termed adversarial energy field optimization for adversarial example purification (AEFOP). AEFOP formulates purification as a constrained optimization problem by defining a learnable adversarial energy which quantifies how far an input deviates from the benign region. This allows adversarial examples to be explicitly pushed from high-energy regions toward low-energy benign regions along an interpretable descent trajectory. Specifically, we build an adversarial energy network and optimize the energy field via a two-stage strategy: adversarial energy field shaping, which enforces distance-like energy behavior and correct gradient directions, and task-driven energy field calibration, which unrolls the descent process to calibrate the field with classification-consistency and semantic-preservation objectives. Extensive experiments across multiple attack scenarios demonstrate that AEFOP achieves superior purification accuracy and high visual quality while requiring only a few gradient steps during inference, offering a practical and efficient robustness layer for vision-based AI services in education. Full article

Magnetic resonance imaging (MRI) is a critical clinical tool that requires precise and reliable interpretation for effective brain tumor diagnosis and timely treatment planning. Deep learning methods have advanced automated tumor classification greatly in the last few years, but many of the current methods are still challenged by a lack of interpretability, a lack of testing on region-focused data, and a lack of model robustness testing. Such limitations reduce clinical trust and limit the practice of automated diagnostic systems. To address these challenges, this study proposes an interpretable deep learning model for classifying brain tumors using the PMRAM dataset, which is a Bangladeshi brain MRI collection containing four categories: glioma, meningioma, pituitary tumor, and normal brain.. The proposed pipeline combines image preprocessing and feature enhancement methods, and then it trains a series of squeeze-and-excitation (SE)-enhanced convolutional neural networks such as VGG19, DenseNet201, MobileNetV3-Large, InceptionV3, and EfficientNetB3. The SE-enhanced EfficientNetB3 performed best, with 98.70% accuracy, 98.77% precision, 98.70% recall, and 98.70% F1-score. Cross-validation also demonstrated stable performance, with a mean accuracy of 96.89%. The model also exhibited efficient inference with low GPU memory consumption, enabling predictions in about 2–4 s per MRI image. Grad-CAM++ and saliency maps were used to improve the transparency of the results, and it was found that the network was concentrated on the clinically significant parts of the tumor, which affected the model predictions. Further robustness analysis and cross-dataset testing are additional evidence of the generalization possibility of the model. An online application was also implemented to allow real-time prediction and visual explanation of brain tumors. Overall, the proposed framework offers a precise, interpretable, and promising solution to automated brain tumor classification using MRI images. Full article

Diabetic retinopathy is a prevalent condition and a major public health concern due to its detrimental impact on eyesight. Diabetes is a root cause of its development and damages small blood vessels caused by prolonged high blood sugar levels. The degenerative consequences of diabetic retinopathy are irrevocable if not diagnosed in the early stages of its progression. This ailment triggers the development of retinal lesions, which can be identified for diagnosis and prognosis. However, lesion detection is challenging due to their similarity in intensity profiles to other retinal features, inconsistent sizes, and random locations. This research evaluates a custom deep learning network for classifying retinal images and compares it with the state-of-the-art classifiers. The novel preprocessing method is introduced to reduce the complexity of the diagnostic process and to enhance classification performance by adaptively enhancing images. Despite being a shallow network, the proposed model yields competitive results with an accuracy of 87.66% and an F1-score of 0.78. The evaluation metrics indicate that class imbalance affects the performance of the proposed model despite using the weighted cross-entropy loss. The future contribution will be the inclusion of generative adversarial networks for generating synthetic images to balance the dataset. This research aims to develop a robust computer-aided diagnostic system as a second interpreter for ophthalmologists during the diagnosis and prognosis stages. Full article