Search Results (223)

Objectives: The increasing use of fetal MRI has increased the diagnosis of posterior fossa malformations, yet the long-term neurodevelopmental outcomes of affected fetuses remain unclear. This study aims to examine the long-term neurodevelopmental outcomes of fetuses with structural posterior fossa malformation diagnosed on fetal MRI. Methods: A historical cohort study was conducted at a single tertiary referral center, including fetuses diagnosed with structural posterior fossa malformations and apparently healthy fetuses who underwent fetal brain MRI between 2011 and 2019. Maternal, pregnancy, and newborn characteristics were compared between groups, alongside long-term neurodevelopmental outcomes using the Vineland Adaptive Behavior Scales II (VABS-II) questionnaire. This included an extensive assessment of malformation types, additional structural, genetic, or neurodevelopmental anomalies, and outcomes. Results: A total of 126 fetuses met the inclusion criteria, of which 70 were apparently healthy fetuses, and 56 had structural posterior fossa malformations. Among the latter, 18 pregnancies were terminated, 4 resulted in neonatal death, and 11 were lost to follow-up. No significant differences were found in the overall neurodevelopmental outcomes between fetuses with structural posterior fossa malformation (93.4 ± 19.0) and apparently healthy fetuses (99.8 ± 13.8). Motor skills scores were lower among fetuses with structural posterior fossa malformations (87.7 ± 16.5 vs. 99.3 ± 17.2, p = 0.01) but remained within the normal range. Conclusion: Fetuses with structural posterior fossa malformations may exhibit normal long-term neurodevelopmental outcomes if no additional anomalies are detected during thorough prenatal screening that includes proper sonographic, biochemical and genetic screening, as well as fetal MRI. Further research with larger cohorts and longer-term assessments is recommended to validate these findings and support clinical decision-making. Full article

Background: Identifying the etiological factors of syringomyelia, which can cause progressive neurological deficits in the spinal cord, is critically important for both diagnosis and treatment. This study aimed to assess the cranial morphometric features of patients with idiopathic syringomyelia by conducting comparative analyses with individuals diagnosed with Chiari Type I, Chiari Type I accompanied by syringomyelia, and healthy controls, in order to elucidate the potential structural contributors to the pathogenesis of idiopathic syringomyelia. Methods: In this retrospective and comparative study, a total of 172 patients diagnosed with Chiari Type I and/or syringomyelia between 2016 and 2024, along with 156 radiologically normal individuals, were included. The participants were categorized into four groups: healthy controls, Chiari Type I, Chiari Type I with syringomyelia, and idiopathic syringomyelia (defined as syringomyelia without an identifiable cause). Midline sagittal T1-weighted MR images were used to obtain quantitative measurements of the posterior fossa, cerebellum, intracranial area, and foramen magnum. All measurements were stratified and statistically analyzed by sex. Results: In cases with idiopathic syringomyelia, both the posterior fossa area and the cerebellum/posterior fossa ratio differed significantly from those of healthy controls. In male patients, the foramen magnum diameter was significantly larger in the Chiari + syringomyelia group compared with the idiopathic group. A significant correlation was found between the degree of tonsillar descent and selected morphometric parameters in female subjects, whereas no such correlation was observed in males. Both Chiari groups exhibited significantly smaller posterior fossa dimensions compared with the healthy and idiopathic groups, indicating greater neural crowding. Additionally, in Chiari Type I patients, increasing degrees of tonsillar descent were associated with a decreased incidence of syringomyelia. Conclusions: Anatomical variations such as a reduced posterior fossa area or altered foramen magnum diameter may contribute to the pathogenesis of idiopathic syringomyelia. Cranial morphometric analysis appears to offer diagnostic value in these cases. Further prospective, multicenter studies incorporating advanced neuroimaging modalities, particularly those assessing cerebrospinal fluid dynamics, are warranted to better understand the mechanisms underlying syringomyelia of unknown etiology. Full article

Accurate segmentation of cardiac MR images using deep neural networks is crucial for cardiac disease diagnosis and treatment planning, as it provides quantitative insights into heart anatomy and function. However, achieving high segmentation accuracy relies heavily on extensive, precisely annotated datasets, which are costly and time-consuming to obtain. This study addresses this challenge by proposing a novel data augmentation framework based on a condition-guided diffusion generative model, controlled by multiple cardiac labels. The framework aims to expand annotated cardiac MR datasets and significantly improve the performance of downstream cardiac segmentation tasks. The proposed generative data augmentation framework operates in two stages. First, a Label Diffusion Module is trained to unconditionally generate realistic multi-category spatial masks (encompassing regions such as the left ventricle, interventricular septum, and right ventricle) conforming to anatomical prior probabilities derived from noise. Second, cardiac MR images are generated conditioned on these semantic masks, ensuring a precise one-to-one mapping between synthetic labels and images through the integration of a spatially-adaptive normalization (SPADE) module for structural constraint during conditional model training. The effectiveness of this augmentation strategy is demonstrated using the U-Net model for segmentation on the enhanced 2D cardiac image dataset derived from the M&M Challenge. Results indicate that the proposed method effectively increases dataset sample numbers and significantly improves cardiac segmentation accuracy, achieving a 5% to 10% higher Dice Similarity Coefficient (DSC) compared to traditional data augmentation methods. Experiments further reveal a strong correlation between image generation quality and augmentation effectiveness. This framework offers a robust solution for data scarcity in cardiac image analysis, directly benefiting clinical applications. Full article

Radar echo extrapolation is a critical forecasting tool in the field of meteorology, playing an especially vital role in nowcasting and weather modification operations. In recent years, spatiotemporal sequence prediction models based on deep learning have garnered significant attention and achieved notable progress in radar echo extrapolation. However, most of these extrapolation network architectures are built upon convolutional neural networks, using radar echo images as input. Typically, radar echo intensity values ranging from −5 to 70 dBZ with a resolution of 5 dBZ are converted into 0–255 grayscale images from pseudo-color representations, which inevitably results in the loss of important echo details. Furthermore, as the extrapolation time increases, the smoothing effect inherent to convolution operations leads to increasingly blurred predictions. To address the algorithmic limitations of deep learning-based echo extrapolation models, this study introduces three major improvements: (1) A Deep Convolutional Generative Adversarial Network (DCGAN) is integrated into the ConvLSTM-based extrapolation model to construct a DCGAN-enhanced architecture, significantly improving the quality of radar echo extrapolation; (2) Considering that the evolution of radar echoes is closely related to the surrounding meteorological environment, the study incorporates specific physical variable products from the initial zero-hour field of RMAPS-NOW (the Rapid-update Multiscale Analysis and Prediction System—NOWcasting subsystem), developed by the Institute of Urban Meteorology, China. These variables are encoded jointly with high-resolution (0.5 dB) radar mosaic data to form multiple radar cells as input. A multi-channel radar echo extrapolation network architecture (MR-DCGAN) is then designed based on the DCGAN framework; (3) Since radar echo decay becomes more prominent over longer extrapolation horizons, this study departs from previous approaches that use a single model to extrapolate 120 min. Instead, it customizes time-specific loss functions for spatiotemporal attenuation correction and independently trains 20 separate models to achieve the full 120 min extrapolation. The dataset consists of radar composite reflectivity mosaics over North China within the range of 116.10–117.50°E and 37.77–38.77°N, collected from June to September during 2018–2022. A total of 39,000 data samples were matched with the initial zero-hour fields from RMAPS-NOW, with 80% (31,200 samples) used for training and 20% (7800 samples) for testing. Based on the ConvLSTM and the proposed MR-DCGAN architecture, 20 extrapolation models were trained using four different input encoding strategies. The models were evaluated using the Critical Success Index (CSI), Probability of Detection (POD), and False Alarm Ratio (FAR). Compared to the baseline ConvLSTM-based extrapolation model without physical variables, the models trained with the MR-DCGAN architecture achieved, on average, 18.59%, 8.76%, and 11.28% higher CSI values, 19.46%, 19.21%, and 19.18% higher POD values, and 19.85%, 11.48%, and 9.88% lower FAR values under the 20 dBZ, 30 dBZ, and 35 dBZ reflectivity thresholds, respectively. Among all tested configurations, the model that incorporated three physical variables—relative humidity (rh), u-wind, and v-wind—demonstrated the best overall performance across various thresholds, with CSI and POD values improving by an average of 16.75% and 24.75%, respectively, and FAR reduced by 15.36%. Moreover, the SSIM of the MR-DCGAN models demonstrates a more gradual decline and maintains higher overall values, indicating superior capability in preserving echo structural features. Meanwhile, the comparative experiments demonstrate that the MR-DCGAN (u, v + rh) model outperforms the MR-ConvLSTM (u, v + rh) model in terms of evaluation metrics. In summary, the model trained with the MR-DCGAN architecture effectively enhances the accuracy of radar echo extrapolation. Full article

Deep learning-based individual recognition of beef cattle has improved the efficiency and effectiveness of individual recognition, providing technical support for modern large-scale farms. However, issues such as over-reliance on back patterns, similar patterns of adjacent cattle leading to low recognition accuracy, and difficulties in deploying models on edge devices exist in the process of group cattle recognition. In this study, we proposed a model based on improved YOLO v5. Specifically, a Simple, Parameter-Free (SimAM) attention module is connected with the residual network and Multidimensional Collaborative Attention mechanism (MCA) to obtain the MCA-SimAM-Resnet (MRS-ATT) module, enhancing the model’s feature extraction and expression capabilities. Then, the $L_{MPDIoU}$ loss function is used to improve the localization accuracy of bounding boxes during target detection. Finally, structural pruning is applied to the model to achieve a lightweight version of the improved YOLO v5. Using 211 test images, the improved YOLO v5 model achieved an individual recognition precision (P) of 93.2%, recall (R) of 94.6%, mean Average Precision (mAP) of 94.5%, FLOPs of 7.84, 13.22 M parameters, and an average inference speed of 0.0746 s. The improved YOLO v5 model can accurately and quickly identify individuals within groups of cattle, with fewer parameters, making it easy to deploy on edge devices, thereby accelerating the development of intelligent cattle farming. Full article

Accurate imaging methods are important for understanding electrodeposition phenomena in metal batteries. Among the suitable imaging methods for this task is magnetic resonance imaging (MRI), which is a very powerful radiological diagnostic method. In this study, MR microscopy was used to image electroplating in a lithium symmetric cell, which was used as a model for a lithium-metal battery. Lithium electrodeposition in this cell was studied by sequential 3D ¹H MRI of 1 M LiPF₆ in EC/DMC electrolyte under different charging conditions, which resulted in different dynamics of the amount of electroplated lithium and its structure. The acquired images depicted the electrolyte distribution, so that the images of deposited lithium that did not give a detectable signal corresponded to the negatives of these images. With this indirect MRI, phenomena such as the transition from a mossy to a dendritic structure at Sand’s time, the growth of whiskers, the growth of dendrites with arborescent structure, the formation of dead lithium, and the formation of gas due to electrolyte decomposition were observed. In addition, the effect of charge and discharge cycles on electrodeposition was also studied. It was found that it is difficult to correctly predict the occurrence of these phenomena based on charging conditions alone, as seemingly identical conditions resulted in different results. Full article

Background: Acute ischemic stroke (AIS) remains a leading cause of mortality and disability worldwide. Accurate evaluation of collateral circulation is essential for predicting outcomes following endovascular thrombectomy (EVT). However, conventional visual collateral scoring (vCS) based on multiphase CT angiography (mCTA) is limited by subjectivity and inter-observer variability. This preliminary study introduces the multiphase quantitative collateral score (mqCS), a novel imaging biomarker designed to provide an objective and reproducible assessment of both the morphological extent and temporal dynamics of collateral flow. Methods: In this exploratory study, 54 AIS patients treated with EVT were retrospectively analyzed. Collateral status was evaluated using both vCS (graded by two blinded neuroradiologists) and mqCS, derived from mCTA-based fractal dimension (FD) and delay indicator (DI) metrics. Logistic regression and receiver operating characteristic (ROC) analyses were performed to assess the predictive value of each scoring system for favorable 90-day functional outcomes (modified Rankin scale, mRS ≤ 2). Results: The mqCS was significantly associated with favorable outcomes. Patients with mqCS ≥ 0.8674 had significantly higher odds of achieving favorable outcomes (adjusted OR = 5.98, 95% CI: 1.38–25.93, p = 0.017; AUC = 0.80). In comparison, the visual collateral score (vCS) showed a lower adjusted predictive value (adjusted OR = 2.84, 95% CI: 1.17–6.89, p = 0.02; AUC = 0.79). Patients in the highest mqCS quartiles (Q3–Q4) exhibited significantly better recovery rates (69%, p < 0.01). Conclusions: This proof-of-concept study suggests that mqCS provides a potentially more objective and robust alternative to visual scoring for collateral assessment in AIS. By integrating structural and temporal characteristics, mqCS enhances outcome prediction and may inform EVT decision-making, particularly in borderline cases. These preliminary findings warrant validation in larger, prospective cohorts and support its potential integration into automated imaging platforms. Full article

This study investigates the generation of synthetic CT (sCT) images from zero echo time (ZTE) MRI to support MR-only radiotherapy, which can reduce image registration errors and lower treatment planning costs. Since MRI lacks the electron density data required for accurate dose calculations, generating reliable sCTs is essential. ZTE MRI, offering high bone contrast, was used with two deep learning models: attention deep residual U-Net (ADR-Unet) and derived conditional generative adversarial network (cGAN). Data from 17 head and neck cancer patients were used to train and evaluate the models. ADR-Unet was enhanced with deep residual blocks and attention mechanisms to improve learning and reconstruction quality. Both models were implemented in-house and compared to standard U-Net and Unet++ architectures using image quality metrics, visual inspection, and dosimetric analysis. Volumetric modulated arc therapy (VMAT) planning was performed on both planning CT and generated sCTs. ADR-Unet achieved a mean absolute error of 55.49 HU and a Dice score of 0.86 for bone structures. All the models demonstrated Gamma pass rates above 99.4% and dose deviations within 2–3%, confirming clinical acceptability. These results highlight ADR-Unet and cGAN as promising solutions for accurate sCT generation, enabling effective MR-only radiotherapy. Full article

Background/Objectives: Obstructive salivary gland disorders—primarily sialolithiasis and ductal stenosis—remain a significant source of morbidity, often requiring surgical intervention. Sialendoscopy has emerged as a minimally invasive, gland-preserving technique for both diagnosis and treatment. This retrospective study aimed to evaluate diagnostic and interventional sialendoscopy outcomes in a Romanian patient cohort and to identify gland-specific considerations in the management of salivary gland obstruction; Methods: A total of 89 patients with confirmed obstructive salivary gland disease (parotid or submandibular) were included. The most common indications included lithiasis, ductal stenosis, sialadenitis, and mixed pathologies; two cases of juvenile recurrent parotitis (JRP) were also managed. All patients underwent clinical evaluation, imaging (ultrasound, CBCT, CT, MR sialography), and sialendoscopic treatment between 2021 and 2025 in two centers. Data on demographics, imaging, calculus size, procedural technique, anesthesia, and complications were collected and analyzed using descriptive and inferential statistics; Results: The submandibular gland was more frequently involved (70.8%), with larger calculi compared to the parotid (mean 7.6 mm vs. 5.1 mm; p = 0.004). Minimally invasive techniques were predominantly used: sialolithotomy and intracorporeal lithotripsy were each performed in 32.6% of cases. Submandibulectomy was required in only 5.6% of patients. Most procedures (93.3%) were conducted under local anesthesia. Complication rates were low and primarily minor and self-limiting; Conclusions: Sialendoscopy is a safe and effective gland-preserving approach in managing obstructive salivary gland disorders. Gland-specific anatomy influences diagnostic pathways and therapeutic choices. These findings support broader adoption of sialendoscopy in routine practice and highlight the need for tailored management protocols based on gland involvement and stone characteristics. However, the study is limited by the absence of standardized post-intervention quality-of-life assessments and structured follow-up data on symptom recurrence. Full article

In this study, praseodymium-doped cobalt ferrite nanoparticles (CoFe_2−yPr_yO₄, y = 0–0.2) were synthesized via sol-gel auto-combustion and systematically characterized to assess their structural, morphological, magnetic, and biological properties. X-ray diffraction (XRD) confirmed single-phase cubic cobalt ferrite formation for samples with y ≤ 0.05 and the emergence of a secondary orthorhombic PrFeO₃ phase at higher dopant concentrations. FTIR spectroscopy identified characteristic metal–oxygen vibrations and revealed a progressive shift of absorption bands with increasing praseodymium (Pr) content. Vibrating sample magnetometry (VSM) demonstrated a gradual decline in saturation (Ms) and remanent (Mr) magnetization with Pr doping, an effect further intensified by cyclodextrin surface coating. TEM analyses revealed a particle size increase correlated with dopant level, while SEM images displayed a porous morphology typical of combustion-synthesized ferrites. In vitro cell viability assays showed minimal toxicity in normal human keratinocytes (HaCaT), while significant antiproliferative activity was observed against human cancer cell lines A375 (melanoma), MCF-7 (breast adenocarcinoma), and HT-29 (colorectal adenocarcinoma), particularly in Pr 6-CD and Pr 7-CD samples. These findings suggest that Pr substitution and cyclodextrin coating can effectively modulate the physicochemical and anticancer properties of cobalt ferrite nanoparticles, making them promising candidates for future biomedical applications. Full article

Addressing the challenge of acquiring depth information in aero-engine assembly scenes using monocular vision, which complicates mixed reality (MR) virtual and real occlusion processing, we propose an ORB-SLAM3-based monocular vision assembly scene virtual and real occlusion processing method. The method proposes optimizing ORB-SLAM3 for matching and depth point reconstruction using the MNSTF algorithm. MNSTF can solve the problems of feature point extraction and matching in weakly textured and texture-less scenes by expressing the structure and texture information of the local images. It is then proposed to densify the sparse depth map using the double-three interpolation method, and the complete depth map of the real scene is created by combining the 3D model depth information in the process model. Finally, by comparing the depth values of each pixel point in the real and virtual scene depth maps, the virtual occlusion relationship of the assembly scene is correctly displayed. Experimental validation was performed with an aero-engine piping connector assembly scenario and by comparing it with Holynski’s and Kinect’s methods. The results showed that in terms of virtual and real occlusion accuracy, the average improvement was 2.2 and 3.4 pixel points, respectively. In terms of real-time performance, the real-time frame rate of this paper’s method can reach 42.4 FPS, an improvement of 77.4% and 87.6%, respectively. This shows that the method in this paper has good performance in terms of the accuracy and timeliness of virtual and real occlusion. This study further demonstrates that the proposed method can effectively address the challenges of virtual and real occlusion processing in monocular vision within the context of mixed reality-assisted assembly processes. Full article

Structural health monitoring (SHM) is crucial for ensuring the safety and longevity of military training aircraft, which face demanding conditions such as high maneuverability, variable loads, and extreme environments, leading to structural fatigue. Traditional methods, such as modal analysis, often struggle to handle the multivariate complexity of operational conditions and data variability. Recently, deep learning has emerged as a promising alternative to overcome these limitations. However, deep learning models typically operate in a unidirectional manner, where feedback to the inputs is often neglected. In contrast, biological neurons utilize feedback mechanisms to refine and adapt their responses in natural ecosystems, enabling adaptive learning and error correction. In this context, this study proposes an innovative Convolutional Neural Network with Reversed Mapping (CNN-RM) approach to SHM, which incorporates feedback loops and self-correcting mechanisms. Before feeding the data into CNN-RM, the dataset complexity is reduced through time-series-to-images Continuous Wavelet Transform (CWT), followed by a denoising CNN (DnCNN) to mitigate complex behavior under various conditions. For application, this study utilizes a massive dataset collected from multivariate sensors installed on a decommissioned military training aircraft previously used by the British Royal Air Force and now housed in a laboratory environment. The results revealed that the overall mean of classification metrics for the CNN is 0.9673 (training) and 0.9422 (testing), while for CNN-MR, it is 0.9764 (training) and 0.9515 (testing), showing an improvement of 0.94% in training and 1.00% in testing. These results highlight significant advancements in SHM, recommending the consideration of such learning mechanisms in neural learning models. Full article

Two-dimensional (2D) cine imaging is essential in routine clinical cardiac MR (CMR) exams for assessing cardiac structure and function. Traditional cine imaging requires patients to hold their breath for extended periods and maintain consistent heartbeats for optimal image quality, which can be challenging for those with impaired breath-holding capacity or irregular heart rhythms. This study aims to systematically assess the performance of a deep learning-based reconstruction (Sonic DL Cine, GE HealthCare, Waukesha, WI, USA) for accelerated cardiac cine acquisition. Multiple retrospective experiments were designed and conducted to comprehensively evaluate the technique using data from an MR-dedicated extended cardiac torso anatomical phantom (digital phantom) and healthy volunteers on different cardiac planes. Image quality, spatiotemporal sharpness, and biventricular cardiac function were qualitatively and quantitatively compared between Sonic DL Cine-reconstructed images with various accelerations (4-fold to 12-fold) and fully sampled reference images. Both digital phantom and in vivo experiments demonstrate that Sonic DL Cine can accelerate cine acquisitions by up to 12-fold while preserving comparable SNR, contrast, and spatiotemporal sharpness to fully sampled reference images. Measurements of cardiac function metrics indicate that function measurements from Sonic DL Cine-reconstructed images align well with those from fully sampled reference images. In conclusion, this study demonstrates that Sonic DL Cine is able to reconstruct highly under-sampled (up to 12-fold acceleration) cine datasets while preserving SNR, contrast, spatiotemporal sharpness, and quantification accuracy for cardiac function measurements. It also provides a feasible approach for thoroughly evaluating the deep learning-based method. Full article

Purpose: Accurate and precise navigation in space and postural stability rely on the central integration of multisensory input (vestibular, proprioceptive, visual), weighted according to its reliability, to continuously update the internal estimate of the direction of gravity. In this study, we examined both peripheral and central vestibular functions in a world-renowned 53-year-old male tightrope walker and investigated the extent to which his exceptional performance was reflected in our findings. Methods: Comprehensive assessments were conducted, including semicircular canal function tests (caloric irrigation, rotatory-chair testing, video head impulse testing of all six canals, dynamic visual acuity) and otolith function evaluations (subjective visual vertical, fundus photography, ocular/cervical vestibular-evoked myogenic potentials [oVEMPs/cVEMPs]). Additionally, static and dynamic posturography, as well as video-oculography (smooth-pursuit eye movements, saccades, nystagmus testing), were performed. The participant’s results were compared to established normative values. High-resolution diffusion tensor magnetic resonance imaging (DT-MRI) was utilized to assess motor tract integrity. Results: Semicircular canal testing revealed normal results except for a slightly reduced response to right-sided caloric irrigation (26% asymmetry ratio; cut-off = 25%). Otolith testing, however, showed marked asymmetry in oVEMP amplitudes, confirmed with two devices (37% and 53% weaker on the left side; cut-off = 30%). Bone-conducted cVEMP amplitudes were mildly reduced bilaterally. Posturography, video-oculography, and subjective visual vertical testing were all within normal ranges. Diffusion tensor MRI revealed no structural abnormalities correlating with the observed functional asymmetry. Conclusions: This professional tightrope walker’s exceptional balance skills contrast starkly with significant peripheral vestibular (otolithic) deficits, while MR imaging, including diffusion tensor imaging, remained normal. These findings highlight the critical role of central computational mechanisms in optimizing multisensory input signals and fully compensating for vestibular asymmetries in this unique case. Full article

This work aims to develop a three-dimensional (3D) super-resolution (SR) network that would perform arbitrary-scale 3D whole-heart (WH) magnetic resonance imaging (MRI) super-resolution, while maintaining fine image details. One-hundred-twenty 3D WH MR volumes, acquired using four different sequences, are used in this study for training, validation, and testing. The proposed method utilizes a frequency-domain regularization in training to maintain fine image detail along with a 3D autoencoder framework. It is also trained in manner to enable it to perform arbitrary factor SR. The proposed method is compared against multiple super-resolution algorithms including two state-of-the-art deep learning methods referred to here as ACNS and TFC as well as nearest neighbor interpolation. The proposed method was evaluated quantitatively and compared against the competing methods with the mean result of the proposed method and the improvements provided by the proposed method (reported by mean percentage between the proposed method and all other competing methods) were recorded. The metrics of interest used for the quantitative comparison are peak signal-to-noise ratio (PSNR, mean = 34.10, mean percentage of improvement = 4.5%), structural similarity index measure (SSIM, mean = 0.94, mean percentage of improvement = 2.2%), mean squared error (MSE, mean = 0.00094, mean percentage of improvement = 48.2%), and root mean squared error (RMSE, mean = 0.024, mean percentage of improvement = 31.0%). Moreover, qualitative comparison was performed using multiple visual comparisons. The quantitative results achieved demonstrate that the proposed method regularly outperforms all other comparison methods. The visual comparisons demonstrate that the proposed method outperforms current state-of-the-art methods in preserving fine image details, as well as its ability to do so for multiple SR factors. Full article