Attention U-Net Research Articles

Puzzle sine cosine optimization-based secure communication and brain tumor classification in IoT‐healthcare system

A brain tumor (BT) refers to an irregular accumulation of cells within the brain that proliferates uncontrollably, resulting in the formation of a mass. The accurate classification and early detection are important for effective treatment. In previous researches, the BT exhibited diverse features in terms of size, shape, and location. Moreover, the images used for segmentation, which suffered from image noise, low contrast, and shifting intensities within tissues. These issues are overcome by developing an effective method in this paper named Puzzle Sine Cosine Optimization enabled Deep Kronecker Network (PSCO-DKN) for classifying BT in the Internet of Things (IoT) healthcare system. Firstly, an IoT network is simulated, where the IoT device is used to capture the patient’s Magnetic Resonance Imaging (MRI) images. Further, the images are routed to the Base Station (BS) by employing PSCO. The routing is accomplished by contemplating several fitness parameters including delay, energy, and distance. At the BS, the process for BT classification is implemented as follows. Initially, the pre-processing is done by utilizing the median filter. Afterwards, the segmentation process is done by applying Spatial Attention U-Net (SA-Unet). After that, Statistical features and Shape Local Binary Texture (SLBT) are extracted. At last, BT classification is performed by utilizing the DKN, which is structurally optimized by using PSCO developed by the hybridization of Puzzle Optimization Algorithm (POA) and Sine Cosine Algorithm (SCA). Finally, PSCO-DKN attained superior outcomes of True Negative Rate (TNR) at 90.9 %, True Positive Rate (TPR) at 92.6 %, and accuracy at 87.7 %.

A semantic segmentation framework with UNet-pyramid for landslide prediction using remote sensing data

Landslides are frequent all over the world, posing serious threats to human life, infrastructure, and economic operations, making them chronic disasters. This study proposes a novel landslide detection methodology that is automated and based on a hybrid deep learning approach. Currently, Deep Learning is constrained by the lack of applicability, lack of data, and low efficiency in landslide detection but with recent advancement in deep learning-based solutions for landslide detection has sparked considerable advantages over traditional techniques. In order to prevent and mitigate disaster, we introduced a hybrid model based on remote sensing technologies such as satellite images. Specifically, the proposed approach consists hybrid U-Net model integrated with a pyramid pooling layer for landslide detection, which uses high-resolution landslide images from the Landslide4Sense dataset. The UNet-Pyramid model has the following modifications: To improve feature acquisition and advancements to strengthen the model’s attention U-Net architecture is integrated with the pyramid pooling layers and OBIA technique. The UNet-Pyramid model was trained and validated using labeled images taken from the Landslide4Sense dataset and the validated set using OBIA to improve its efficacy. The overall Precision, Recall, and F1 Score of the UNet-pyramid model for landslide detection are 91%, 84%, and 87%, respectively.

Monitoring the leaf damage by the rice leafroller with deep learning and ultra-light UAV.

Rice leafroller is a serious threat to the production of rice. Monitoring the damage caused by rice leafroller is essential for effective pest management. Owing to limitations in collecting decent quality images and high-performing identification methods to recognize the damage, studies recommending fast and accurate identification of rice leafroller damage are rare. In this study, we employed an ultra-lightweight unmanned aerial vehicle (UAV) to eliminate the influence of the downwash flow field and obtain very high-resolution images of the damaged areas of the rice leafroller. We used deep learning technology and the segmentation model, Attention U-Net, to recognize the damaged area by the rice leafroller. Further, a method is presented to count the damaged patches from the segmented area. The result shows that Attention U-Net achieves high performance, with an F1 score of 0.908. Further analysis indicates that the deep learning model performs better than the traditional image classification method, Random Forest (RF). The traditional method of RF causes a lot of false alarms around the edge of leaves, and is sensitive to the changes in brightness. Validation based on the ground survey indicates that the UAV and deep learning-based method achieve a reasonable accuracy in identifying damage patches, with a coefficient of determination of 0.879. The spatial distribution of the damage is uneven, and the UAV-based image collecting method provides a dense and accurate method to recognize the damaged area. Overall, this study presents a vision to monitor the damage caused by the rice leafroller with ultra-light UAV efficiently. It would also contribute to effectively controlling and managing the hazardous rice leafroller. © 2024 Society of Chemical Industry.

Deep learning-based segmentation of acute ischemic stroke MRI lesions and recurrence prediction within 1 year after discharge: A multicenter study

ObjectiveTo explore the performance of deep learning-based segmentation of infarcted lesions in the brain MRI of patients with AIS and the recurrence prediction value of radiomics within 1 year after discharge as well as to develop a model incorporating radiomics features and clinical factors to accurately predict AIS recurrence. Materials and MethodsTo generate a segmentation model of MRI lesions in AIS, the deep learning algorithm multiscale residual attention UNet (MRA-UNet) was employed. Furthermore, the risk factors for AIS recurrence within 1 year were explored using logistic regression analysis. In addition, to develop the prediction model for AIS recurrence within 1 year after discharge, four machine learning algorithms, namely, LR, RandomForest, CatBoost, and XGBoost, were employed based on radiomics data, clinical data, and their combined data. ResultsIn the validation set, the MDice and MIou of the MRA-UNet segmentation model were 0.816 and 0.801, respectively. In multivariate logistic regression analysis, age, renal insufficiency, C-reactive protein, triglyceride glucose index, prognostic nutritional index, and infarct volume were identified as the independent risk factors for stroke recurrence. Furthermore, in the validation set, combining radiomics data and clinical data, the AUC was 0.835 (95%CI:0.738, 0.932), 0.834 (95%CI:0.740, 0.928), 0.858 (95%CI:0.770, 0.946), and 0.842 (95%CI:0.752, 0.932) for the LR, RandomForest, CatBoost, and XGBoost models, respectively. ConclusionThe MRA-UNet model can effectively improve the segmentation accuracy of diffusion-weighted images. The model, which was established by combining radiomics features and clinical factors, held some value for predicting ischemic stroke recurrence within 1 year.

A semi-automatic cardiovascular annotation and quantification toolbox utilizing prior knowledge-guided feature learning

Annotation and quantification of cardiovascular in intravascular ultrasound (IVUS) images are of great significance in assisting the clinical diagnosis and treatment of coronary heart disease. However, it is time-consuming and expensive for cardiologists to manually annotate cardiovascular and calculate parameters in IVUS images using different software during the diagnosis. In this paper, we propose an IVUS analysis toolbox (SaCAQ) for semi-automatic cardiovascular annotation and quantification via prior knowledge-guided feature learning. The SaCAQ consists of two main modules: (1) a semi-automatic cardiovascular annotation module and (2) a cardiovascular quantification module. The semi-automatic cardiovascular annotation module can accurately segment the vascular wall by a IVUS images segmentation model (ISM-IVUS). The ISM-IVUS first extracts vascular morphological features and pixel intensity change features for the complex image segmentation by the expert prior knowledge feature extraction mechanism. Secondly, vascular wall is accurately segmented by the multi-scale feature extraction of attention U-Net. Moreover, the segmentation result checking mechanism is applied to verify the automatic segmentation result, which enhances the reliability of SaCAQ. Plaque regions of interest to doctors are detected by the rule-based plaque detection method, which can assist in the quantification task. Finally, image measurement methods and medically recognized calculation formulas are used in the cardiovascular quantification module to obtain clinical parameters, which significantly improves the confidence of the results. The SaCAQ is verified on our own dataset containing 5242 IVUS slice images and 383 sets of parameters. The experiments demonstrate that the SaCAQ achieves high accuracy of vascular wall segmentation: mIoU of 86.16%, Dice coefficient of 93.43%, and Hausdorff distance of 0.2171 mm. Moreover, ISM-IVUS has good segmentation speed and robustness. The calculated parameters show good correlation and p-values. All of these results indicate that the proposed SaCAQ provides an efficient, accurate, and reliable toolbox for clinical diagnosis of cardiovascular disease. The SaCAQ and user manual can be publicly available from https://github.com/zHwiz/SaCAQ.

Cross-shaped windows transformer with self-supervised pretraining for clinically significant prostate cancer detection in bi-parametric MRI.

Bi-parametric magnetic resonance imaging (bpMRI) has demonstrated promising results in prostate cancer (PCa) detection. Vision transformers have achieved competitive performance compared to convolutional neural network (CNN) in deep learning, but they need abundant annotated data for training. Self-supervised learning can effectively leverage unlabeled data to extract useful semantic representations without annotation and its associated costs. This study proposes a novel self-supervised learning framework and a transformer model to enhance PCa detection using prostate bpMRI. We introduce a novel end-to-end Cross-Shaped windows (CSwin) transformer UNet model, CSwin UNet, to detect clinically significant prostate cancer (csPCa) in prostate bpMRI. We also propose a multitask self-supervised learning framework to leverage unlabeled data and improve network generalizability. Using a large prostate bpMRI dataset (PI-CAI) with 1476 patients, we first pretrain CSwin transformer using multitask self-supervised learning to improve data-efficiency and network generalizability. We then finetune using lesion annotations to perform csPCa detection. We also test the network generalization using a separate bpMRI dataset with 158 patients (Prostate158). Five-fold cross validation shows that self-supervised CSwin UNet achieves 0.888±0.010 aread under receiver operating characterstics curve (AUC) and 0.545±0.060 Average Precision (AP) on PI-CAI dataset, significantly outperforming four comparable models (nnFormer, Swin UNETR, DynUNet, Attention UNet, UNet). On model generalizability, self-supervised CSwin UNet achieves 0.79 AUC and 0.45 AP, still outperforming all other comparable methods and demonstrating good generalization to external data. This study proposes CSwin UNet, a new transformer-based model for end-to-end detection of csPCa, enhanced by self-supervised pretraining to enhance network generalizability. We employ an automatic weighted loss (AWL) to unify pretext tasks, improving representation learning. Evaluated on two multi-institutional public datasets, our method surpasses existing methods in detection metrics and demonstrates good generalization to external data.

Training ESRGAN with multi-scale attention U-Net discriminator

In this paper, we propose MSA-ESRGAN, a novel super-resolution model designed to enhance the perceptual quality of images. The key innovation of our approach lies in the integration of a multi-scale attention U-Net discriminator, which allows for more accurate differentiation between subject and background areas in images. By leveraging this architecture, MSA-ESRGAN surpasses traditional methods and several state-of-the-art super-resolution models in terms of Natural Image Quality Evaluator (NIQE) scores as well as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) across various benchmark datasets, including BSD100, Set5, Set14, Urban100, and OST300. Additionally, subjective evaluations further confirm the enhanced visual quality delivered by MSA-ESRGAN, particularly in terms of preserving texture and overall image realism. To ensure a fair comparison with Real-ESRGAN, we initialized our generator with a pre-trained Real-ESRNET model and followed the same training setup. Our model was trained on the DIV2K dataset using high-resolution image patches and the Adam optimizer, incorporating exponential moving average (EMA) for stability and performance enhancement. Evaluations on multiple benchmark datasets demonstrate that MSA-ESRGAN consistently delivers superior perceptual quality, as evidenced by higher NIQE, PSNR, and SSIM scores compared to other methods. Specifically, our model shows significant improvements in both objective and subjective measures of image quality. Furthermore, an ablation study highlighted the critical role of our multi-scale attention U-Net discriminator in enhancing the model’s performance. The results underscore the effectiveness of MSA-ESRGAN in maintaining image naturalness and perceptual quality, providing a robust benchmark for blind super-resolution tasks.

Deep learning applied to the segmentation of rodent brain MRI data outperforms noisy ground truth on full-fledged brain atlases

Translational magnetic resonance imaging of the rodent brain provides invaluable information for preclinical drug development. However, the automated segmentation of such images for quantitative analyses is limited compared to human brain imaging mainly due to the inferior anatomical contrast and the resulting less advanced registration and atlasing tools. Here, we investigated the potential of deep learning models for the segmentation of magnetic resonance images of rat brains into an entire set of multiple regions of interest (rather than individual loci), focusing on the development of a robust method that accommodates changes in the input based on differences in animal strain (genotype) and size. Manually generated labels are expensive, so we tested the ability of neural networks to learn brain structures from noisy but inexpensive registration-based labels, allowing very large datasets to be leveraged for training. We compared three distinct model architectures (U-Net, Attention-U-Net and DeepLab) by training them on a dataset of >10,000 magnetic resonance images of rat brains and found that each model was able to segment the entire brain into predefined sets of 29 and 58 regions, respectively, with the Attention U-Net achieving the best performance. The models canceled out unstructured label noise in the imperfect training data to provide smoother and more symmetric segmentations than registration-based labeling, and were more robust when presented with input variations, thus outperforming the noisy ground truth. Our pipeline also includes uncertainty estimation and an explainability mechanism, hence providing features essential for anomaly detection and quality assurance. In summary, our study shows that deep learning models do achieve accurate brain segmentation in high-throughput quantitative preclinical imaging without the need for expensive expert-generated labels.

Resolution-dependent MRI-to-CT translation for orthotopic breast cancer models using deep learning

Objective.This study aims to investigate the feasibility of utilizing generative adversarial networks (GANs) to synthesize high-fidelity computed tomography (CT) images from lower-resolution MR images. The goal is to reduce patient exposure to ionizing radiation while maintaining treatment accuracy and accelerating MR image acquisition. The primary focus is to determine the extent to which low-resolution MR images can be utilized to generate high-quality CT images through a systematic study of spatial resolution-dependent magnetic resonance imaging (MRI)-to-CT image conversion.Approach.Paired MRI-CT images were acquired from healthy control and tumor models, generated by injecting MDA-MB-231 and 4T1 tumor cells into the mammary fat pad of nude and BALB/c mice to ensure model diversification. To explore various MRI resolutions, we downscaled the highest-resolution MR image into three lower resolutions. Using a customized U-Net model, we automated region of interest masking for both MRI and CT modalities with precise alignment, achieved through three-dimensional affine paired MRI-CT registrations. Then our customized models, Nested U-Net GAN and Attention U-Net GAN, were employed to translate low-resolution MR images into high-resolution CT images, followed by evaluation with separate testing datasets.Main Results.Our approach successfully generated high-quality CT images (0.142mm2) from both lower-resolution (0.282mm2) and higher-resolution (0.142mm2) MR images, with no statistically significant differences between them, effectively doubling the speed of MR image acquisition. Our customized GANs successfully preserved anatomical details, addressing the typical loss issue seen in other MRI-CT translation techniques across all resolutions of MR image inputs.Significance.This study demonstrates the potential of using low-resolution MR images to generate high-quality CT images, thereby reducing radiation exposure and expediting MRI acquisition while maintaining accuracy for radiotherapy.

Flood Mapping Through Satellite Images Using Deep Learning

Flooding is a pervasive natural disaster that significantly impacts agriculture, infrastructure, and communities, particularly in rural and agricultural regions. Traditional methods of flood assessment and compensation are often manual, time-intensive, and prone to inaccuracies, resulting in delayed aid and inequitable resource distribution. With advancements in technology, satellite imagery and artificial intelligence (AI) provide new opportunities for real-time flood mapping and automated disaster response. This project proposes a Real-Time Flood Mapping and Compensation System using the Attention U-Net deep learning model. Satellite imagery is analysed to detect and segment flood-affected regions with high precision. The Attention U-Net architecture enhances segmentation accuracy by focusing on the most relevant regions in the images while ignoring irrelevant or noisy areas, such as clouds or non-flooded landscapes. The system integrates geospatial data on land boundaries to assess flood severity for individual properties, enabling fair and data-driven compensation for affected landowners. This project holds significant potential for transforming flood disaster management by reducing response times, fostering transparency, and empowering affected communities to recover more effectively. The integration of advanced AI and geospatial technologies marks a critical step toward mitigating the impacts of natural disasters in a rapidly changing climate

Remote Sensing Identification of Picea schrenkiana var. tianschanica in GF-1 Images Based on a Multiple Mixed Attention U-Net Model

Remote sensing technology plays an important role in woodland identification. However, in mountainous areas with complex terrain, accurate extraction of woodland boundary information still faces challenges. To address this problem, this paper proposes a multiple mixed attention U-Net (MMA-U-Net) semantic segmentation model using 2015 and 2022 GF-1 PMS images as data sources to improve the ability to extract the boundary features of Picea schrenkiana var. tianschanica forest. The U-Net architecture serves as its underlying network, and the feature extraction ability of the Picea schrenkiana var. tianschanica is improved by adding hybrid attention CBAM and replacing the original skip connection with the DCA module to improve the accuracy of the model segmentation. The results show that on the remote sensing dataset with GF-1 PMS images, compared with the original U-Net and other models, the accuracy of the multiple mixed attention U-Net model is increased by 5.42%–19.84%. By statistically analyzing the spatial distribution of Picea schrenkiana var. tianschanica as well as their changes, the area was 3471.38 km2 in 2015 and 3726.10 km2 in 2022. Combining the predicted results with the DEM data, it was found that the Picea schrenkiana var. tianschanica were most distributed at an altitude of 1700–2500 m. The method proposed in this study can accurately identify Picea schrenkiana var. tianschanica and provides a theoretical basis and research direction for forest monitoring.

Comparative Assessment of Different Architectures of Convolutional Neural Network for Semantic Segmentation of Forest Disturbances from Multi-Temporal Satellite Images

Algorithms based on convolutional neural networks are the most efficient for semantic segmentation of images, including segmentation of forest cover disturbances from satellite images. In this study, we consider the applicability of various modifications of the U-net architecture of convolutional neural network for recognizing logged, burnt and windthrow areas in forests from multi-temporal and multi-seasonal Sentinel-2 satellite images. The assessment was carried out on three test sites that differ significantly in the characteristics of forest stands and forest management. The highest accuracy (average F-measure of 0.59) was obtained from the U-net model, while the models that showed the best results during training (Attention U-Net and MobilNetv2 U-Net) did not improve segmentation of independent data. The resulting accuracy estimates are close to those previously published for forests with a substantial proportion of selective logged areas. Characteristics of logged areas and windthrows, namely their area and type are the main factor determining the accuracy of semantic segmentation. Substantial differences were also revealed between the images taken in different seasons of the year, with the maximum segmentation accuracy based on winter pairs of images. According to summertime and different-season pairs of images, the area of forest disturbances is substantially underestimated. Forest species composition has a less significant effect, although for two of the three test sites, the maximum accuracy was observed in dark coniferous forests, and the minimum in deciduous forests. There was no statistically significant effect of slope lighting factor calculated from digital elevation model on segmentation accuracy based for winter pairs of images. The accuracy of segmentation of burnt areas, which was assessed using the example of 14 large forest fires in 2021-2022, is unsatisfactory, which is probably due to the varying degrees of damage to the forest cover in the burnt areas.

A Novel Multi-Scale Feature Enhancement U-Shaped Network for Pixel-Level Road Crack Segmentation

Timely and accurate detection of pavement cracks, the most common type of road damage, is essential for ensuring road safety. Automatic image segmentation of cracks can accurately locate their pixel positions. This paper proposes a Multi-Scale Feature Enhanced U-shaped Network (MFE-UNet) for pavement crack detection. This network model uses a Residual Detail-Enhanced Block (RDEB) instead of a conventional convolution in the encoder–decoder process. The block combines Efficient Multi-Scale Attention to enhance its feature extraction performance. The Multi-Scale Gating Feature Fusion (MGFF) is incorporated into the skip connections, enhancing the fusion of multi-scale features to capture finer crack details while maintaining rich semantic information. Furthermore, we created a pavement crack image dataset named China_MCrack, consisting of 1500 images collected from road surfaces using smartphone-mounted motorbikes. The proposed network was trained and tested on the China_MCrack, DeepCrack, and Crack-Forest datasets, with additional generalization experiments on the BochumCrackDataset. The results were compared with those of the U-Net model, ResUNet, and Attention U-Net. The experimental results show that the proposed MFE-UNet model achieves accuracies of 82.95%, 91.71%, and 69.02% on three datasets, namely, China_MCrack, DeepCrack, and Crack-Forest datasets, respectively, and the F1_score is improved by 1–4% compared with other networks. Experimental results demonstrate that the proposed method is effective in detecting cracks at the pixel level.

SEGMENTATION OF THYROID NODULES ON ULTRASOUND IMAGES

The increasing prevalence of thyroid cancer in our country and globally has led to the development of various computer-aided studies for its detection, contributing significantly to the literature. Artificial intelligence and image processing are particularly prominent methods in this field due to their non-invasive nature, accessibility, and ability to provide valuable information about the morphological characteristics of nodules. In recent years, segmentation algorithms in medical imaging have garnered substantial interest for their potential to enhance diagnostic accuracy. Accurate segmentation of thyroid nodules is a critical first step in the development of AI-assisted clinical decision support systems for the detection and diagnosis of thyroid cancer. In this study, innovative methods were employed to detect thyroid nodules. A dice score of 79% was achieved in instance segmentation using the YOLOv5-Small algorithm when doppler images were excluded, while a dice score of 91% was obtained using the YOLOv5-Large algorithm on a dataset that included doppler images. In semantic segmentation, the Attention Unet++ and Manet algorithms achieved a dice score of 89% when doppler images were excluded, and 91% when they were included. These results demonstrate that images typically excluded by physicians could potentially offer better outcomes in computerized image processing.

Demystifying the effect of receptive field size in U-Net models for medical image segmentation.

Medical image segmentation is a critical task in healthcare applications, and U-Nets have demonstrated promising results in this domain. We delve into the understudied aspect of receptive field (RF) size and its impact on the U-Net and attention U-Net architectures used for medical imaging segmentation. We explore several critical elements including the relationship among RF size, characteristics of the region of interest, and model performance, as well as the balance between RF size and computational costs for U-Net and attention U-Net methods for different datasets. We also propose a mathematical notation for representing the theoretical receptive field (TRF) of a given layer in a network and propose two new metrics, namely, the effective receptive field (ERF) rate and the object rate, to quantify the fraction of significantly contributing pixels within the ERF against the TRF area and assessing the relative size of the segmentation object compared with the TRF size, respectively. The results demonstrate that there exists an optimal TRF size that successfully strikes a balance between capturing a wider global context and maintaining computational efficiency, thereby optimizing model performance. Interestingly, a distinct correlation is observed between the data complexity and the required TRF size; segmentation based solely on contrast achieved peak performance even with smaller TRF sizes, whereas more complex segmentation tasks necessitated larger TRFs. Attention U-Net models consistently outperformed their U-Net counterparts, highlighting the value of attention mechanisms regardless of TRF size. These insights present an invaluable resource for developing more efficient U-Net-based architectures for medical imaging and pave the way for future exploration of other segmentation architectures. A tool is also developed, which calculates the TRF for a U-Net (and attention U-Net) model and also suggests an appropriate TRF size for a given model and dataset.

Exploring rooftop solar potential through deep learning: a comparative case study of U-Net and Attention U-Net

Photovoltaic systems are increasingly gaining popularity for rooftop installations in urban areas. These devices contribute to the self-sufficiency of electricity supply and reduce greenhouse gas emissions in urban settings. In this study, we implement deep learning techniques and compare the U-Net and Attention U-Net architectures to detect rooftops in the city of Sidi Bel Abbes (North Africa) for the installation of solar panels. Contrary to trends observed in the literature and previous research, our results show that U-Net achieved a higher accuracy of 90.49%, while Attention U-Net achieved an accuracy of 89.96%. Using the U-Net method, we identified a 1.9-square-kilometer area of rooftops suitable for solar panel installation in the southern part of the town of Sidi Bel Abbes. This discovery represents a potential electrical capacity of 89 MVA, capable of largely satisfying the energy needs of the studied region. This work is a novelty, being the first study of its kind undertaken in Algeria.

U-Net and Its Variants Based Automatic Tracking of Radial Artery in Ultrasonic Short-Axis Views: A Pilot Study.

Background/Objectives: Radial artery tracking (RAT) in the short-axis view is a pivotal step for ultrasound-guided radial artery catheterization (RAC), which is widely employed in various clinical settings. To eliminate disparities and lay the foundations for automated procedures, a pilot study was conducted to explore the feasibility of U-Net and its variants in automatic RAT. Methods: Approved by the institutional ethics committee, patients as potential RAC candidates were enrolled, and the radial arteries were continuously scanned by B-mode ultrasonography. All acquired videos were processed into standardized images, and randomly divided into training, validation, and test sets in an 8:1:1 ratio. Deep learning models, including U-Net and its variants, such as Attention U-Net, UNet++, Res-UNet, TransUNet, and UNeXt, were utilized for automatic RAT. The performance of the deep learning architectures was assessed using loss functions, dice similarity coefficient (DSC), and Jaccard similarity coefficient (JSC). Performance differences were analyzed using the Kruskal-Wallis test. Results: The independent datasets comprised 7233 images extracted from 178 videos of 135 patients (53.3% women; mean age: 41.6 years). Consistent convergence of loss functions between the training and validation sets was achieved for all models except Attention U-Net. Res-UNet emerged as the optimal architecture in terms of DSC and JSC (93.14% and 87.93%), indicating a significant improvement compared to U-Net (91.79% vs. 86.19%, p < 0.05) and Attention U-Net (91.20% vs. 85.02%, p < 0.05). Conclusions: This pilot study validates the feasibility of U-Net and its variants in automatic RAT, highlighting the predominant performance of Res-UNet among the evaluated architectures.

Marine Oil Pollution Monitoring Based on a Morphological Attention U-Net Using SAR Images.

This study proposed an improved full-scale aggregated MobileUNet (FA-MobileUNet) model to achieve more complete detection results of oil spill areas using synthetic aperture radar (SAR) images. The convolutional block attention module (CBAM) in the FA-MobileUNet was modified based on morphological concepts. By introducing the morphological attention module (MAM), the improved FA-MobileUNet model can reduce the fragments and holes in the detection results, providing complete oil spill areas which were more suitable for describing the location and scope of oil pollution incidents. In addition, to overcome the inherent category imbalance of the dataset, label smoothing was applied in model training to reduce the model's overconfidence in majority class samples while improving the model's generalization ability. The detection performance of the improved FA-MobileUNet model reached an mIoU (mean intersection over union) of 84.55%, which was 17.15% higher than that of the original U-Net model. The effectiveness of the proposed model was then verified using the oil pollution incidents that significantly impacted Taiwan's marine environment. Experimental results showed that the extent of the detected oil spill was consistent with the oil pollution area recorded in the incident reports.

An improved convolutional architecture for quantitative characterization of pore networks in fine-grained rocks using FIB-SEM

Focused ion beam scanning electron microscope (FIB-SEM) is one of the most advanced imaging techniques for analyzing and understanding complex pore networks in shale and other fine-grained formations. However, FIB-SEM imaging tends to be time-consuming and labor-intensive and can result in biased interpretations associated with pore analysis. Recently, U-Net or its variants for image segmentation have been applied to capture microscopic pores at higher resolutions. The ‘traditional’ encoder-decoder-based approaches tend to detect very fine-scale microscopic pores poorly. This study presents an improved convolutional architecture for automatically analyzing pore structures in shale reservoirs using FIB-SEM. It does so by applying an overcomplete convolutional architecture, KiU-Net, to capture very fine-scale microscopic pores by accurately defining their edges in the input FIB-SEM images. The KiU-Net learns low and high-level features by making the model more sensitive to fine-scale microscopic pores in the input images. The purpose of this study is to demonstrate KiU-Net's capabilities by analyzing different shale formations with varying characteristics. The results indicate that KiU-Net is more accurate and efficient than other methods in predicting nanopores in the Longmaxi, Niutitang, Qingshankou, Qianjiang, and Yanchang Formations (China), Bakken shale (Canada), and coal reservoirs (China). Furthermore, KiU-Net demonstrated the advantage of requiring fewer parameters and achieving super convergence compared to the Attention U-Net technique. KiU-Net addresses the challenges of the Edge-Threshold Automatic Processing (ETAP) methods by capturing very fine-scale microscopic pores with accurate edges. This study further enhances the accuracy and efficiency of pore analysis in shales, thereby offering an improved method for understanding shale reservoir quality with the potential to improve petroleum recovery from such formations.

A novel 8-connected Pixel Identity GAN with Neutrosophic (ECP-IGANN) for missing imputation

Missing pixel imputation presents a critical challenge in image processing and computer vision, particularly in applications such as image restoration and inpainting. The primary objective of this paper is to accurately estimate and reconstruct missing pixel values to restore complete visual information. This paper introduces a novel model called the Enhanced Connected Pixel Identity GAN with Neutrosophic (ECP-IGANN), which is designed to address two fundamental issues inherent in existing GAN architectures for missing pixel generation: (1) mode collapse, which leads to a lack of diversity in generated pixels, and (2) the preservation of pixel integrity within the reconstructed images. ECP-IGANN incorporates two key innovations to improve missing pixel imputation. First, an identity block is integrated into the generation process to facilitate the retention of existing pixel values and ensure consistency. Second, the model calculates the values of the 8-connected neighbouring pixels around each missing pixel, thereby enhancing the coherence and integrity of the imputed pixels. The efficacy of ECP-IGANN was rigorously evaluated through extensive experimentation across five diverse datasets: BigGAN-ImageNet, the 2024 Medical Imaging Challenge Dataset, the Autonomous Vehicles Dataset, the 2024 Satellite Imagery Dataset, and the Fashion and Apparel Dataset 2024. These experiments assessed the model’s performance in terms of diversity, pixel imputation accuracy, and mode collapse mitigation, with results demonstrating significant improvements in the Inception Score (IS) and Fréchet Inception Distance (FID). ECP-IGANN markedly enhanced image segmentation performance in the validation phase across all datasets. Key metrics, such as Dice Score, Accuracy, Precision, and Recall, were improved substantially for various segmentation models, including Spatial Attention U-Net, Dense U-Net, and Residual Attention U-Net. For example, in the 2024 Medical Imaging Challenge Dataset, the Residual Attention U-Net’s Dice Score increased from 0.84 to 0.90, while accuracy improved from 0.88 to 0.93 following the application of ECP-IGANN. Similar performance enhancements were observed with the other datasets, highlighting the model’s robust generalizability across diverse imaging domains.