Search Results (851)

Background: Non-arteritic anterior ischemic optic neuropathy (NAION) is a neuroophthalmological disorder characterized by impaired blood flow to the optic nerve head. There is uncertainty about whether, in some cases, NAION may be caused by proximal embolism of the posterior ciliary arteries. Diffusion-weighted magnetic resonance imaging (DWI-MRI) can provide evidence of concurrent cerebral infarction that may indicate a common embolic etiology. Methods: Adults with ophthalmological diagnosis of NAION who underwent cerebral DWI-MRI within 14 days from onset of visual impairment were included in a retrospective cohort study (2013–2021). DWI-MRI images were assessed for presence, location, and type of ischemic stroke by a board-certified neuroradiologist blinded for clinical patient data. Results: Among 122 patients (mean age 64.6 ± 11.9 years), DWI-MRI indicated acute/subacute ischemic stroke in three cases (2.5%), all located within the anterior circulation in the territory of the left middle cerebral artery and ipsilateral to the affected eye in two cases (1.6%). Ischemic stroke location was cortical in one case (0.8%) and subcortical in two cases (1.6%). Acute ischemic stroke indicated by a hyperintense DWI signal and corresponding low ADC was present in one patient (0.8%). Two patients (1.6%) had subacute ischemic stroke (hyperintense DWI signal and normal or elevated ADC signal). Only one NAION patient (0.8%) had acute embolic stroke corresponding to the vascular territory of the affected eye. Conclusions: Concurrent embolic ischemic stroke in NAION is exceedingly rare. Our findings support the prevailing pathophysiological theory of NAION as a non-embolic disease. Full article

All-sky imagers (ASIs) used in solar energy meteorology face an extreme intra-image dynamic range, with the circumsolar neighborhood orders of magnitude brighter than the diffuse dome. Many operational ASI pipelines address this gap with high-dynamic-range (HDR) bracketing inside the camera’s image signal processor (ISP), i.e., after demosaicing and color processing in a nonlinear 8-bit RGB domain. Near the Sun, such ISP-domain HDR can down-weight the shortest exposure, retain clipped or near-clipped samples from longer frames, and compress highlight contrast, thereby increasing circumsolar saturation and flattening aureole gradients. A radiance-linear HDR fusion in the sensor/RAW domain (RAW–HDR) is therefore contrasted with the vendor ISP-based HDR mode (ISP–HDR). Solar-based geometric calibration enables Sun-centered analysis. Paired, interleaved acquisitions under clear-sky and broken-cloud conditions are evaluated using two circumsolar performance criteria per RGB channel: (i) saturated-area fraction in concentric rings and (ii) a median-based radial gradient in defined arcs. All quantitative analyses operate on the radiance-linear HDR result; post-merge tone mapping is only used for visualization. Across conditions, ISP–HDR exhibits roughly double the near-saturation within 0–$4^{°}$ of the Sun and about a three- to fourfold weaker circumsolar radial gradient within 0–$6^{°}$ relative to RAW–HDR. These findings indicate that radiance-linear fusion in the RAW domain better preserves circumsolar structure than the examined ISP-domain HDR mode and thus provides more suitable input for downstream tasks such as cloud–edge detection, aerosol retrieval, and irradiance estimation. Full article

Background/Objectives: This study aimed to evaluate the diagnostic accuracy of Apparent Diffusion Coefficient (ADC) values, derived from Diffusion-Weighted Imaging (DWI), in differentiating benign and malignant ascites. Methods: This retrospective study included 150 patients (85 benign, 65 malignant) who underwent abdominal MRI. All patients were scanned on a DWI sequence (b-values: 0, 500, and 1000 s/mm²). Two experienced radiologists, blinded to clinical and cytological outcomes, measured the mean ADC (ADCmean) from three distinct ROIs and the minimum ADC (ADCmin) from the area of lowest signal intensity on the ADC map. The diagnostic performance of ADC parameters and the Serum-Ascites Albumin Gradient (SAAG) was assessed using Receiver Operating Characteristic (ROC) curve analysis. Results: The mean values of ADCmean (3162 ± 204 × 10⁻⁶ mm²/s) and ADCmin (2885 ± 148 × 10⁻⁶ mm²/s) in the malignant group were significantly lower than those in the benign group (3596 ± 239 and 3322 ± 218 × 10⁻⁶ mm²/s; p = 0.006 and p = 0.0016, respectively). Inter-observer agreement was good for both ADCmean (ICC = 0.844) and ADCmin (ICC = 0.879). In the ROC analysis, ADCmin demonstrated the highest diagnostic performance (AUC: 0.930). An optimal cut-off value for ADCmin of ≤ 2983 × 10⁻⁶ mm²/s yielded 81.5% sensitivity and 85.8% specificity. The diagnostic performance of ADCmin was found to be superior to that of ADCmean (AUC: 0.877) and SAAG (AUC: 0.919). Conclusions: ADC values derived from DWI, particularly ADCmin, represent a highly accurate, non-invasive, and reproducible biomarker for differentiating benign from malignant ascites. The identified ADCmin threshold provides quantitative parameter that can aid in patient triage, especially when cytology is inconclusive, potential surrogate for fluid characterization. Full article

Hyperspectral images (HSIs) provide rich spectral–spatial information and support applications in remote sensing, agriculture, and medicine, yet their development is hindered by data scarcity and costly acquisition. Diffusion models have enabled synthetic HSI generation, but conventional integer-order solvers such as Denoising Diffusion Implicit Models (DDIM) and Pseudo Linear Multi-Step method (PLMS) require many steps and rely mainly on local information, causing error accumulation, spectral distortion, and inefficiency. To address these challenges, we propose Integer–Fractional Alternating Diffusion Sampling (IFADiff), a training-free inference-stage enhancement method based on an integer–fractional alternating time-stepping strategy. IFADiff combines integer-order prediction, which provides stable progression, with fractional-order correction that incorporates historical states through decaying weights to capture long-range dependencies and enhance spatial detail. This design suppresses noise accumulation, reduces spectral drift, and preserves texture fidelity. Experiments on hyperspectral synthesis datasets show that IFADiff consistently improves both reference-based and no-reference metrics across solvers without retraining. Ablation studies further demonstrate that the fractional order $\alpha$ acts as a controllable parameter: larger values enhance fine-grained textures, whereas smaller values yield smoother results. Overall, IFADiff provides an efficient, generalizable, and controllable framework for high-quality HSI generation, with strong potential for large-scale and real-time applications. Full article

Sparse unmixing (SU) has become a research hotspot in hyperspectral image (HSI) analysis in recent years due to its interpretable physical mechanisms and engineering practicality. However, traditional SU methods are confronted with two core bottlenecks: Firstly, the high computational complexity of the abundance matrix inversion severely limits algorithmic efficiency. Secondly, the inherent challenges posed by large-scale highly coherent spectral libraries hinder improvement of unmixing accuracy. To overcome these limitations, this study proposes a novel sub-abundance map regularized sparse unmixing (SARSU) framework based on dynamic abundances subspace awareness. Specifically, first of all, we have developed an intelligent spectral atom selection strategy that employs a designed dynamic activity evaluation mechanism to quantify the participation contribution of spectral library atoms during the unmixing process in real time. This enables adaptive selection of critical subsets to construct active subspace abundance maps, effectively mitigating spectral redundancy interference. Secondly, we innovatively integrated weighted nuclear norm regularization based on sub-abundance maps into the model, deeply mining potential low-rank structures within spatial distribution patterns to significantly enhance the spatial fidelity of unmixing results. Additionally, a multi-directional neighborhood-aware dual total variation (DTV) regularizer was designed, which enforces spatial consistency constraints between adjacent pixels through a four directional (horizontal, vertical, diagonal, and back-diagonal) differential penalty mechanism, ensuring abundance distributions comply with physical diffusion laws of ground objects. Finally, to efficiently solve the proposed objective model, an optimization algorithm based on the Alternating Direction Method of Multipliers (ADMM) was developed. Comparative experiments conducted on two simulated datasets and four real hyperspectral benchmark datasets, alongside comparisons with state-of-the-art methods, validated the efficiency and superiority of the proposed approach. Full article

Background: Preoperative breast MRI is widely used in surgical planning because of its high sensitivity. However, discrepancies with histopathology remain common and can affect tumor size assessment and treatment decisions. In addition, recent comparative studies have highlighted the growing role of contrast-enhanced mammography (CEM) and multiparametric MRI (mpMRI), both of which may improve specificity and accessibility compared to conventional MRI. Methods: A structured literature review was conducted in PubMed (2000–2025) according to PRISMA guidelines. Studies included if they evaluated preoperative breast MRI with histopathological correlation and reported sensitivity, specificity, or concordance outcomes. Data extraction focused on study design, patient and tumor characteristics, imaging methods, and clinical impact. Results: MRI demonstrates high sensitivity, particularly in detecting IDC and ILC. However, overestimation of tumor size remains a concern, particularly in ILC and high-grade DCIS, while underestimation is frequently observed after neoadjuvant therapy, especially in Luminal A tumors. Tumor size and stage significantly affect concordance, with advanced-stage tumors (T2–T3) showing better MRI-histopathology concordance than early-stage lesions (T0–T1). Specificity remains limited, particularly in DCIS and multifocal disease. Emerging evidence suggests that contrast-enhanced mammography (CEM) achieves comparable sensitivity with higher specificity, while multiparametric MRI (mpMRI) incorporating diffusion-weighted imaging (DWI) improves lesion characterization and prediction of treatment response. Conclusions: While MRI remains a valuable diagnostic tool for breast cancer, histopathological validation is essential to guide treatment decisions. Future research should focus on AI-enhanced imaging techniques, CEM and multiparametric MRI to improve concordance rates, reduce overdiagnosis and translate imaging advances into meaningful clinical outcomes. Full article

Background/Objectives: This study explored the associations of normal breast tissue characteristics on multiparametric MRI with clinical assessments of breast cancer risk in women with dense breasts. Methods: Women with dense breasts who underwent multiparametric MRI were included. Breast cancer risk was determined based on Tyrer–Cuzick (TC) lifetime risk scores, categorized as high (TC ≥ 20%) or low risk. Qualitative background parenchymal enhancement (BPE) assessment was obtained from imaging reports. Quantitative imaging markers were calculated, including median BPE, median apparent diffusion coefficient, and volume measures of the whole breast, fibroglandular tissue (FGT), blood vessels, and BPE regions. The associations between imaging markers and TC risk groups were evaluated using age-adjusted logistic regression and summarized by area under the receiver operating characteristic curve (AUC). Results: Seventy-seven women were evaluated; a total of 20 (26%) were low risk, and 57 (74%) were high risk. After adjusting for age and multiple testing, BPE:breast ratio (adj. p = 0.037), FGT:breast ratio (adj. p = 0.046), and BPE:vessel ratio (adj. p = 0.037) were positively associated with risk, while qualitative BPE was not (adj. p = 0.11). Overall, risk categorizations based on imaging markers were concordant with TC score in up to 70% of women. Conclusions: In women with dense breasts, quantitative measures from multiparametric MRI (BPE:breast, FGT:breast, and BPE:vessel ratios) moderately discriminated high- and low-risk groups, warranting further investigation of their value to supplement conventional breast cancer risk assessment tools. Full article

Small-cell lung cancer (SCLC) is an aggressive malignancy that exhibits high prevalence for brain metastases. Furthermore, chemotherapy and metastasis-preventive approaches are also linked to neurotoxicity, further aggravating cognitive impairment. Despite evidence supporting structural and functional brain alterations in SCLC, the application of machine learning (ML) to new connectivity biomarkers has remained unexplored. This study is—to the best of our knowledge—the first to apply ML to structural brain connectomics in SCLC, using diffusion tensor imaging (DTI) to identify features discriminating between post-chemotherapy SCLC patients and healthy controls. Specifically, we constructed structural networks via deterministic tractography, applying an adapted feature reduction technique to identify the most informative connections without selection bias. This process isolated 16 connections involving 26 brain regions, predominantly in the frontal, temporal, and parietal lobes, showcasing primarily intra-hemispheric and left-lateralized alterations. Our optimal model leveraged a Gaussian Support Vector Machine (SVM), achieving a weighted accuracy of 0.92, a sensitivity of 0.93, a specificity of 0.91, and an area under the curve of 0.94. The selected feature subset retained high performance when tested with other classifiers, confirming its robustness. Our findings differ from prior studies based on statistically derived features, highlighting the ML-driven connectomics’ potential in uncovering DTI-derived SCLC patterns, offering interpretable insights for neuroimaging-based diagnostics. Full article

Background/Objectives: Superior cerebellar artery (SCA) infarction is a rare but clinically significant subtype of posterior circulation stroke. Extensive swelling in the SCA territory may cause downward brainstem compression and appear as brainstem hypodensity on computed tomography, potentially leading to premature treatment withdrawal. Methods: We report the case of a 50-year-old woman with acute SCA-territory infarction (NIHSS = 7) presenting with vertigo, dysphagia, dysarthria, and diplopia. Initial computed tomography suggested extensive brainstem infarction, prompting withdrawal of treatment. Diffusion-weighted MRI revealed reversible edema with brainstem sparing. The patient underwent suboccipital decompressive craniectomy and ventricular drainage with favorable neurological recovery. In addition, a systematic literature search was conducted according to PRISMA 2020 guidelines in PubMed, Web of Science, and Scopus (studies published since 1 January 2015). Fifteen studies met predefined eligibility criteria. Results: Magnetic resonance imaging findings were decisive in avoiding a falsely dismal prognosis and inappropriate withdrawal of care. Across the literature, infarct volume (>30–35 mL), brainstem involvement and bilateral cerebellar infarction emerged as key predictors of malignant course. Early decompressive surgery was consistently associated with improved survival, though functional outcomes varied. Fast magnetic resonance imaging techniques and volumetric imaging improved risk stratification and surgical decision-making. Conclusions: SCA infarction can mimic brainstem infarction on computed tomography due to secondary compression rather than true ischemia. Magnetic resonance imaging is essential to guide treatment and prevent avoidable mortality. Multimodal imaging combined with interdisciplinary management allows for accurate prognostication and optimized surgical timing in malignant SCA infarction. Full article

To explore the relationship between magnetic resonance imaging (MRI) parameters, including T2-weighted intensity and apparent diffusion coefficient (ADC), and intraoperative tumor characteristics, particularly consistency, in vestibular schwannomas (VSs). The association between tumor consistency, facial nerve (FN) function, and postoperative outcomes was analyzed. A single-center retrospective analysis included newly diagnosed VS cases (2020–2023) with cisternal involvement (Samii T3a; volume ≥ 0.7 cm³). T2 and ADC maps from the perimetral region of interest were normalized, and tumors were categorized into 3 classes by combining qualitative consistency (soft, fibrous, or fibrous/hard), ultrasonic aspirator power, and adherence to neurovascular structures. FN function was assessed using the House–Brackmann scale at the immediate postoperative period and 12-month follow-up. MRIs of 33 VSs (18 solid and 15 cystic) were analyzed. Normalized values of both T2 (N-T2_mean) and ADC (N-ADC_min) maps predicted the classical radiological differentiation. N-ADC_min may have some role in predicting consistency (value 1.361, p = 0.017, accuracy 0.48) and demonstrated a significant association (p = 0.04) with the FN outcome in the immediate postoperative period. An augmented consistency could impair FN function by increasing the intrameatal pressure related to greater transmission of shocks derived from the dissection maneuvers of the cisternal component of the tumor. The possibility of non-invasively exploring VS consistency with a parameter easily calculable on MRI might be beneficial in surgical planning, modifying the timing of the opening of the meatus with respect to what could be the surgical routine in some centers. Full article

Bamboo, a fast-growing and biodegradable industrial crop, exhibits excellent mechanical properties, which facilitate its widespread use in construction, furniture, and decorative applications. However, its inherently limited permeability hinders processing during drying, chemical modification, dyeing, and impregnation. Although previous studies have explored structural and treatment-related aspects, few have offered a comprehensive and integrative overview that bridges anatomical structure, permeation mechanisms, performance evaluation, and treatment strategies. This review synthesizes 126 publications from 1997 to 2024 to provide a comprehensive, multidimensional analysis of bamboo permeability. Structure–function relationships are examined by assessing how vessels, sieve tubes, perforation plates, pits, and bamboo nodes influence permeability, with an emphasis on quantitative correlations. Capillarity, diffusion, and viscous resistance are integrated into a unified theoretical framework, proposing a model that couples longitudinal capillary rise with transverse diffusion. Detection approaches, including both direct techniques (weight gain, microscopy, tracer elements, fluorescence imaging) and indirect techniques (porosity measurement, Micro-CT), with their respective advantages, limitations, and applications. Enhancement strategies are categorized into chemical, physical, and biological methods, with assessments of their effectiveness, environmental impact, and energy consumption. Overall, this review provides a holistic perspective on bamboo permeability and offers valuable guidance for future research and engineering applications. Full article

Background: Early detection of recurrent brain tumors after malignant glioma surgery is a challenge in imaging-based assessment of glioma. Objective: The aim of this case series is to investigate whether there are signs for an improvement in the early detection of recurrent tumors using multiparametric magnetic resonance imaging (MRI) after glioma surgery. Methods: An MRI protocol was used with high-resolution fluid-attenuated inversion recovery (FLAIR), diffusion tensor imaging (DTI), resting state functional MRI (rsfMRI), and contrast-enhanced high resolution T1-weighted (T1w). Longitudinal multiparametric MRI was performed in six patients with glioblastoma with one complete scan before surgery, one scan after surgery and at least two follow-up scans. A total of 27 complete multiparametric MRI data sets were available. Results: DTI analysis at the localizations of recurrent tumors showed early directionality loss in DTI by fractional anisotropy (FA) reduction accompanied by FLAIR hyperintensities before hyperintensities in contrast enhanced T1w were visible. One out of six patients showed a regional FA decrease at the localization of the recurrent tumor at a point of time even when the morphological T1w- and FLAIR images did not demonstrate any detectable changes. Functional connectivity alterations in a corresponding network could also be detected at the localizations of the recurrent tumor. Conclusions: In addition to routine T2w FLAIR and contrast enhanced T1w, DTI and rsfMRI might complement information for the early detection of recurrent malignant glioma. Prospective studies at larger scale are needed with respect to potential of DTI and rsfMRI for early recurrent tumor detection. Full article

Low-light image enhancement (LLIE) remains a challenging task due to the scarcity of paired training data and the complex signal-dependent noise inherent in low-light scenes. To address these issues, this paper proposes a fully unsupervised framework named Wavelet-Guided Zero-Reference Diffusion (WZD) for natural low-light image restoration. WZD leverages an ImageNet-pre-trained diffusion prior and a multi-scale representation of the Discrete Wavelet Transform (DWT) to restore natural illumination from a single dark image. Specifically, the input low-light image is first processed by a Practical Exposure Corrector (PEC) to provide an initial robust luminance baseline. It is then converted from the RGB to the YCbCr color space. The Y channels of the input image and the current diffusion estimate are decomposed into four orthogonal sub-bands—LL, LH, HL, and HH—and fused via learnable, step-wise weights while preserving structural integrity. An exposure control loss and a detail consistency loss are jointly employed to suppress over/under-exposure and preserve high-frequency details. Unlike recent approaches that rely on complex supervised training or lack physical guidance, our method integrates wavelet guidance with a zero-reference learning framework, incorporates the PEC module as a physical prior, and achieves significant improvements in detail preservation and noise suppression without requiring paired training data. Comprehensive experiments on the LOL-v1, LOL-v2, and LSRW datasets demonstrate that WZD achieves a superior or competitive performance, surpassing all referenced unsupervised methods. Ablation studies confirm the critical roles of the PEC prior, YCbCr conversion, wavelet-guided fusion, and the joint loss function. WZD also enhances the performance of downstream tasks, verifying its practical value. Full article

Background: Locally advanced cervical cancer (LACC) represents a significant challenge in oncology, requiring accurate assessment of local extent and metastatic spread. Multiparametric magnetic resonance imaging (mpMRI) has assumed a central role in the loco-regional characterization of the tumor due to its high soft-tissue resolution and the ability to integrate functional information. Objectives: In this narrative review, we explore the use of mpMRI in the diagnosis, staging, and treatment response of LACC, comparing its performance with that of PET/CT, which remains complementary for remote staging. The potential of whole-body magnetic resonance imaging (WB-MRI) and hybrid PET/MRI techniques is also analyzed, as well as the emerging applications of radiomics and artificial intelligence. The paper also discusses technical limitations, interpretative variability, and the importance of protocol standardization. The goal is to provide an updated and translational summary of imaging in LACC, with implications for clinical practice and future research. Methods: Prospective and retrospective studies, systematic reviews, and meta-analyses on adult patients with cervical cancer were included. Results: Fifty-two studies were included. MRI demonstrated a sensitivity and specificity greater than 80% for parametrial and bladder invasion, but limited sensitivity (45–60%) for lymph node disease, lower than PET/CT. Multiparametric MRI was useful in early prediction of response to chemotherapy and radiotherapy and in distinguishing residual disease from fibrosis. The integration of MRI into Image-Guided Adaptive Brachytherapy (IGABT) resulted in improved oncological outcomes and reduced toxicity. The applications of radiomics and AI demonstrated enormous potential in predicting therapeutic response and lymph node status in the MRI study, but multicenter validation is still needed. Conclusions: MRI is the cornerstone of the local–regional staging of advanced cervical cancer; it has become an essential and crucial tool in treatment planning. Its use, combined with PET/CT for lymph node assessment and metastatic disease staging, is now the standard of care. Future prospects include the use of whole-body MRI and the development of predictive models based on radiomics and artificial intelligence. Full article

UAV object detection is still difficult to achieve due to large-scale variation, dense small objects, a complicated background, and resource constraints from onboard computing. To solve these problems, we develop a diffusion-enhanced detection network, DL-DEIM, tailored for aerial images. The proposed scheme generalizes the DEIM baseline across three orthogonal axes. First, we propose a lightweight backbone network called DCFNet, which utilizes a DRFD module and a FasterC3k2 module to maintain spatial information and reduce computational complexity. Second, we propose a LFDPN module, which can conduct bidirectional multi-scale fusion via frequency-spatial self-attention and deep feature refinement and largely enhance cross-scale contextual propagation for small objects. Third, we propose LAWDown, an adaptive-content-aware downsampling to preserve the discriminative representation with higher accuracy at lower resolutions, which can effectively capture the spatially-variant weights and group channel interactions. On the VisDrone2019 dataset, DL-DEIM achieves a mAP@0.5 of 34.9% and a mAP@0.5:0.95 of 20.0%, outperforming the DEIM baseline by +4.6% and +2.9%, respectively. The model maintains real-time inference speed (356 FPS) with only 4.64 M parameters and 11.73 GFLOPs. Ablation studies validate the fact that DCFNet, LFDPN, and LAWDown collaboratively contribute to the accuracy and efficiency. Visualizations also display clustered and better localized activation in crowded scenes. These results show that DL-DEIM achieves a good tradeoff between detection probability and computation burden and it can be used in practice on resource-limited UAV systems. Full article