Search

The processing and analysis of computed tomography (CT) imaging is important for both basic scientific development and clinical applications. In AutoCT, we provide a comprehensive pipeline that integrates an end-to-end automatic preprocessing, registration, segmentation, and quantitative analysis of 3D CT scans. The engineered pipeline enables atlas-based CT segmentation and quantification leveraging diffeomorphic transformations through efficient forward and inverse mappings. The extracted localized features from the deformation field allow for downstream statistical learning that may facilitate medical diagnostics. On a lightweight and portable software platform, AutoCT provides a new toolkit for the CT imaging community to underpin the deployment of artificial intelligence-driven applications.

Estimates of the land area occupied by wind energy differ by orders of magnitude due to data scarcity and inconsistent methodology. We developed a method that combines machine learning-based imagery analysis and geographic information systems and examined the land area of 318 wind farms (15,871 turbines) in the U.S. portion of the Western Interconnection. We found that prior land use and human modification in the project area are critical for land-use efficiency and land transformation of wind projects. Projects developed in areas with little human modification have a land-use efficiency of 63.8 ± 8.9 W/m2 (mean ±95% confidence interval) and a land transformation of 0.24 ± 0.07 m2/MWh, while values for projects in areas with high human modification are 447 ± 49.4 W/m2 and 0.05 ± 0.01 m2/MWh, respectively. We show that land resources for wind can be quantified consistently with our replicable method, a method that obviates >99% of the workload using machine learning. To quantify the peripheral impact of a turbine, buffered geometry can be used as a proxy for measuring land resources and metrics when a large enough impact radius is assumed (e.g., >4 times the rotor diameter). Our analysis provides a necessary first step toward regionalized impact assessment and improved comparisons of energy alternatives.

The work presents a newly designed penalty function to be added with an existing Cross Entropy based fitness function [3] for optimal selection of multi-level thresholds for image segmentation. The extended fitness function so designed is tested here for Brain magnetic resonance (MR) image segmentation using nature-inspired meta-heuristics such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Whale Optimization Algorithm (WOA) and Moth-flame Optimization (MFO), and MFO is finally selected. The proposed method excels in Brain MR image segmentation when tested on WBA database and BrainWeb MR image database with other nature-inspired meta-heuritic based methods. It outperformed the other methods in terms of the three performance metrics Peak Signal to Noise Ratio (PSNR), Structural Similarity Index (SSIM) and Feature Similarity Index (FSIM). The best results shown by the proposed method on WBA database in terms of these metrics - PSNR, SSIM and FSIM - are 29.77, 0.927 and 0.899 respectively. On BrainWeb database, the proposed method yields 26.91, 0.864 and 0.892 for the three respective metrics. [ABSTRACT FROM AUTHOR]

In this study, we redefine FCM algorithm by integrating fuzzy set theory, fuzzy metrics, and Sugeno negation principles. This innovative approach overcomes the limitations inherent in conventional machine learning models, especially in situations characterized by uncertainty, noise, and ambiguity. Our model utilizes the membership degrees from fuzzy set theory, and transforms the concept of proximity defined by fuzzy metrics into a minimization problem. This transformation is achieved using a linguistic negation operator, which is crucial for optimizing FCM algorithm's objective function. A significant innovation in our research is the use of GA for optimizing parameters within the contexts of fuzzy metrics and Sugeno negation. The precise optimization capabilities of GA greatly enhance the sensitivity and adaptability of FCM algorithm, thereby improving overall performance. By leveraging the meticulous parameter adjustments provided by GA, our approach has shown superior results in practical applications, such as brain MRI image segmentation, surpassing traditional methods. Experimental results highlight the considerable enhancements our proposed FCM algorithms bring over existing methods across various performance metrics. In conclusion, this study makes a valuable addition to the field of fuzzy-based machine learning methodologies. It combines the optimization strength of GA with the flexible classification capabilities of fuzzy logic. The integration of Sugeno negation and fuzzy metrics not only improves the accuracy and precision of FCM algorithm but also provides significant benefits in handling complex and ambiguous datasets. This research signifies a major advance in machine learning and fuzzy logic, setting the stage for future applications and studies. [ABSTRACT FROM AUTHOR]

Recent years have witnessed notable advancements in medical image segmentation through deep convolutional neural networks. However, a notable limitation lies in the local operation of convolution, which hinders the ability to fully exploit global semantic information. To overcome the challenges prevalent in medical image segmentation, the feature ensemble network with multi‐scale atrous transformer is proposed. At the core of the approach lies the multi‐scale contextual integration module, which is based on the multi‐scale atrous transformer and facilitates contextual integration of multi‐level features. To extract discriminative fine‐grained features of the target region, a hybrid attention mechanism that synergistically combines spatial and channel attention, thereby sharpening the model's focus on crucial target information within high‐level features, is incorporated. Additionally, the channel‐aware feature reconstruction module is introduced as an innovative component engineered to tackle feature similarity issues across different categories. This module performs feature reconstruction based on channel perception, effectively widening the feature gap between categories and enhancing the segmentation capability. It is worth mentioning that our approach surpasses the state‐of‐the‐art method using three benchmark datasets in medical image segmentation. [ABSTRACT FROM AUTHOR]

Traditional video steganography methods primarily rely on modifying concealed spaces for embedding, thereby exhibiting a certain degree of security and embedding capacity. Nevertheless, these methods do not fully capitalize on the rich semantic information inherent in videos, limiting their overall effectiveness. In this paper, an AI‐generated video steganography scheme based on semantic segmentation is proposed. The mapping relationship between secret and semantic information is established by using a semantic segmentation model. The secret information can be converted into semantic labels by semantic histograms or pixels means, and semantic labels containing secret information are obtained and input into the video‐to‐video model to drive the generation of stego videos. After receiving the stego video, the receiver extracts the secret information using a pre‐defined specific embedding mode, including the methods of sub‐block partitioning and embedding capacity per frame. The experimental results show that the stego video has good visual quality, security, and robustness against various noise attacks. [ABSTRACT FROM AUTHOR]

Effective extraction of building from remote sensing images requires both global and local information. Despite convolutional neural networks (CNNs) excelling at capturing local details, their intrinsic focus on local operations poses challenge in effectively extracting global features, especially in the context of large‐scale buildings. In contrast, transformers excel at capturing global information, but compared to CNNs, they tend to overly rely on large‐scale datasets and pre‐trained parameters. To tackle the challenge, this paper presents the multi‐scale and dual‐branch feature fusion network (MDTrans). Specifically, the CNN and transformer branches are integrated in a dual‐branch parallel manner during both encoding and decoding stages, local information for small‐scale buildings is extracted by utilizing Dense Connection Blocks in the CNN branch, while crucial global information for large‐scale buildings is effectively captured through Swin Transformer Block in the transformer branch. Additionally, Dual Branch Information Fusion Block is designed to fuse local and global features from the two branches. Furthermore, Multi‐Convolutional Block is designed to further enhance the feature extraction capability of buildings with different sizes. Through extensive experiments on the WHU, Massachusetts, and Inria building datasets, MDTrans achieves intersection over union (IoU) scores of 91.36%, 64.69%, and 79.25%, respectively, outperforming other state‐of‐the‐art models. [ABSTRACT FROM AUTHOR]

Automatic segmentation of renal tumours during renal cell carcinoma surgery can help doctors accurately locate the tumour region, protect the tissues and organs around the kidneys, enhance surgical efficiency, and reduce the possibility of leakage and misdiagnosis. However, since general polyp endoscopic image segmentation models have many problems when facing the task of renal tumour segmentation, there needs to be more research on the segmentation of endoscopic images of renal tumours. This paper proposes a multi‐scale dynamic segmentation kernel network for endoscopic image segmentation of kidney tumours. First, a spatial receptive field module is proposed to augment the feature information and improve the performance of the whole network. Second, an enhanced cross‐attention module is offered to attenuate the effect of a high‐similarity segmentation background. Finally, a multi‐scale dynamic segmentation kernel module is introduced to gradually refine the segmentation results from small to large sizes to obtain more accurate tumour boundaries. Extensive experiments on the established kidney tumour endoscopic dataset and publicly available endoscopic datasets show that this method exhibits enhanced performance and generalization capabilities compared to existing techniques. On this renal tumour dataset, MDSK‐Net achieved excellent results of 94.1% and 90.1% on mDice and mIoU. [ABSTRACT FROM AUTHOR]

Additive manufacturing (AM) applications are rapidly expanding across multiple domains and are not limited to prototyping purposes. However, achieving flawless parts in medical, aerospace, and automotive applications is critical for the widespread adoption of AM in these industries. Since AM is a complex process consisting of multiple interdependent factors, deep learning (DL) approaches are adopted widely to correlate the AM process physics to the part quality. Typically, in AM processes, computer vision-based DL is performed by extracting the machine's sensor data and layer-wise images through camera-based systems. This paper presents an overview of computer vision-assisted patch-wise defect localization and pixel-wise segmentation methods reported for AM processes to achieve error-free parts. In particular, these deep learning methods localize and segment defects in each layer, such as porosity, melt-pool regions, and spattering, during in situ processes. Further, knowledge of these defects can provide an in-depth understanding of fine-tuning optimal process parameters and part quality through real-time feedback. In addition to DL architectures to identify defects, we report on applications of DL extended to adjust the AM process variables in closed-loop feedback systems. Although several studies have investigated deploying closed-loop systems in AM for defect mitigation, specific challenges exist due to the relationship between inter-dependent process parameters and hardware constraints. We discuss potential opportunities to mitigate these challenges, including advanced segmentation algorithms, vision transformers, data diversity for improved performance, and predictive feedback approaches. [ABSTRACT FROM AUTHOR]

Numerical simulation studies of water–rock interaction mechanisms and pore-scale multiphase flow properties often require high computational efficiency and realistic geometries to enable a fast and accurate description of hydrodynamic behavior. In this paper, we have chosen to use deep learning models to achieve these requirements, firstly by using encoder structures to refine the image segmentation of void-solid structures on complex geometries of scanning electron microscopy (SEM) images of porous media through few-shot learning (FSL), not only obtaining an accuracy of 0.97, but also reducing the amount of annotation work. We then focus on pore-scale three-dimensional (3D) structural reconstruction using the unpaired image-to-image translation method, optimizing the cycle-consistent adversarial network (cycle-GAN) model via sliced Wasserstein distance (SWD) to transfer marine sedimentary sandstone features to the initial image, and the geometric stochastic reconstruction problems are transformed into optimization problems. Subsequently, the computational efficiency was improved by a factor of 21 by implementing the lattice Boltzmann simulation method (LBM) accelerated by GPU through compute-unified device architecture (CUDA). The flow field distribution and absolute permeability of the extracted 2D samples and the reconstructed 3D porous media structure were simulated. The results showed that our method could rapidly and accurately reconstruct the 3D structures of a given feature, ensuring statistical equivalence between the 3D reconstructed structures and 2D samples. We solve the problem of extrapolation-based 3D reconstruction of porous media and significantly reduce the time spent on structure extraction and numerical calculations. [ABSTRACT FROM AUTHOR]

The CoAtNet deep neural model has been shown to achieve state-of-the-art performance by stacking convolutional and self-attention layers. In particular, the initial layers of CoAtNet apply efficient convolutions for extracting local features out of the input image and the initial fine-resolution feature maps. In turn, the final layers apply more cumbersome Transformers in order to extract global features from the coarse-resolution feature maps. The model's outcome directly depends on those final global features. This paper proposes an extension of the original CoAtNet model based on the introduction of a dual stream of convolution and self-attention blocks applied at the final layers of CoAtNet. In this way, those final layers automatically aggregate both local and global features extracted from the initial feature maps. Two dual-stream topologies have been proposed and evaluated. This Dual-Stream CoAtNet model exhibits a significant improvement on the segmentation accuracy of breast ultrasound images, thus contributing to the development of more robust tumor detection methods. [ABSTRACT FROM AUTHOR]

In order to better handle images with intensity inhomogeneity and noise, an adaptive multi-level set active contour model based on block search is proposed in this paper. This model first defines a multiple edge extension criterion for the input image to block the image and avoid the loss of image edge information; Then, the proposed adaptive block search method is used to find the level set that is considered redundant, and the remaining parts are fused to obtain a rough binary mask; Finally, the target is extracted by using the newly defined energy functional and the edge contours of the extracted binary mask. The experimental results show that the average jaccard similarity coefficients of the proposed model for segmenting images with intensity inhomogeneity and real images are 97.81% and 98.41%, respectively, and the accuracy of the segmentation results is higher than that of other models participating in the comparison. Similarly, the results of the ablation experiment once again validated the robustness of the proposed model. [ABSTRACT FROM AUTHOR]

Deformable image registration (DIR) plays an important part in many clinical tasks, and deep learning has made significant progress in DIR over the past few years.To propose a fast multiscale unsupervised deformable image registration (referred to as FMIRNet) method for monomodal image registration.We designed a multiscale fusion module to estimate the large displacement field by combining and refining the deformation fields of three scales. The spatial attention mechanism was employed in our fusion module to weight the displacement field pixel by pixel. Except mean square error (MSE), we additionally added structural similarity (ssim) measure during the training phase to enhance the structural consistency between the deformed images and the fixed images.Our registration method was evaluated on EchoNet, CHAOS and SLIVER, and had indeed performance improvement in terms of SSIM, NCC and NMI scores. Furthermore, we integrated the FMIRNet into the segmentation network (FCN, UNet) to boost the segmentation task on a dataset with few manual annotations in our joint leaning frameworks. The experimental results indicated that the joint segmentation methods had performance improvement in terms of Dice, HD and ASSD scores.Our proposed FMIRNet is effective for large deformation estimation, and its registration capability is generalizable and robust in joint registration and segmentation frameworks to generate reliable labels for training segmentation tasks. [ABSTRACT FROM AUTHOR]

Stroke, the second leading cause of mortality globally, predominantly results from ischemic conditions. Immediate attention and diagnosis, related to the characterization of brain lesions, play a crucial role in patient prognosis. Standard stroke protocols include an initial evaluation from a non-contrast CT to discriminate between hemorrhage and ischemia. However, non-contrast CTs lack sensitivity in detecting subtle ischemic changes in this phase. Alternatively, diffusion-weighted MRI studies provide enhanced capabilities, yet are constrained by limited availability and higher costs. Hence, we idealize new approaches that integrate ADC stroke lesion findings into CT, to enhance the analysis and accelerate stroke patient management. This study details a public challenge where scientists applied top computational strategies to delineate stroke lesions on CT scans, utilizing paired ADC information. Also, it constitutes the first effort to build a paired dataset with NCCT and ADC studies of acute ischemic stroke patients. Submitted algorithms were validated with respect to the references of two expert radiologists. The best achieved Dice score was 0.2 over a test study with 36 patient studies. Despite all the teams employing specialized deep learning tools, results reveal limitations of computational approaches to support the segmentation of small lesions with heterogeneous density. [ABSTRACT FROM AUTHOR]

Fingerprint recognition systems hold a pivotal role across various modern applications, such as criminal investigation, civil identification, and access control. Fingerprint segmentation is commonly the first stage of fingerprint recognition systems, focusing on extracting the foreground from captured fingerprint images. This helps to reduce extracting false minutiae, speed up the extraction process, and hence, improve the overall system performance. In this paper, we introduce a novel fingerprint segmentation method by first proposing new frequency and intensity features with employing fuzzy-c-means and genetic algorithm, accompanied by proposing a new cluster assignment strategy that involves features weighting and cluster assignment probabilities thresholding which helps to improve the accuracy compared to the standard cluster assignment method. Finally, we introduce a new morphological post-processing technique in order to minimize the misclassified pixels and obtain the final mask. Furthermore, we carry out a comprehensive performance evaluation against leading fingerprint segmentation methods on various known databases. As a result, the proposed method achieves an average error rate of 2.52%, which is much lower than the corresponding errors obtained in the literature. By these experimental results, we show that the proposed method outperforms the best existing methods. [ABSTRACT FROM AUTHOR]

Skin disease image segmentation faces two major challenges: the complex and varied lesion morphology and the presence of interfering image backgrounds. To address these difficulties in skin disease image segmentation, we propose a Residual U-Net architecture with Channel-Space Separate Attention based on depthwise separable convolutions. The multi-scale residual U-Net modules in the encoder efficiently capture multi-scale texture information in lesions and backgrounds within a single stage, overcoming the limitations of U-Net in extracting just local features. The introduction of ConvMixer Block for global contextual modeling contributes to suppress complex background interference and enhances the overall understanding of lesion morphology. Additionally, we employ a Channel-Space Separate Attention mechanism with depthwise separable convolutions(CSSA-DSC) for feature fusion, effectively addressing the limited expressiveness issue associated with U-Net’s direct skip-connection concatenation. Experimental results on the PH2, ISIC 2017, and ISIC 2018 datasets demonstrate our method’s strong multi-scale modeling and feature expression capabilities. [ABSTRACT FROM AUTHOR]

Deep learning techniques have attained substantial progress in various face‐related tasks, such as face recognition, face inpainting, and facial expression recognition. To prevent infection or the spread of the virus, wearing of masks in public places has been mandated following the COVID‐19 epidemic, which has led to face occlusion and posed significant challenges for face recognition systems. Most prominent masked face recognition solutions rely on mask segmentation tasks. Therefore, segmentation can be used to mitigate the negative impacts of wearing a mask and improve recognition accuracy. Mask region segmentation suffers from two main problems: there is no standard type of masks that people wear, they come in different colours and designs, and there is no publicly available masked face dataset with appropriate ground truth for the mask region. In order to address these issues, we propose an encoder–decoder framework that utilizes a boundary‐aware attention network combined with a new hybrid loss to provide a map, patch, and pixel‐level supervision. We also introduce a dataset called MFSD, with 11,601 images and 12,758 masked faces for masked face segmentation. Furthermore, we compare the performance of different cutting‐edge deep learning semantic segmentation models on the presented dataset. Experimental results on the MSFD dataset reveal that the suggested approach outperforms state‐of‐the‐art, algorithms with 97.623% accuracy, 93.814% IoU, and 96.817% F1‐score rate. Our dataset of masked faces with mask region labels and source code will be available online. [ABSTRACT FROM AUTHOR]

Medical image segmentation through the use of deep learning is becoming a trend targeting to automate disease detection and provide effective treatment. Traditionally huge manual efforts are associated with medical image processing by medical staff. The wide use of neural networks driven by the progressive advancement in computing power and the availability of training data provides instrumental means to automate medical image processing. A polyp refers to an abnormal growth that can occur in various parts of the body. While the majority of polyps are noncancerous or benign, there are instances where certain types can develop into cancer. Detecting and segmenting polyps is highly valuable for identifying early signs and potentially enabling more effective treatment of colon cancer. In this paper, we introduce a novel method for segmenting polyp images. The proposed method utilizes convolutional neural networks (CNNs) and employs an enhanced attention mechanism. To cater to a more granular attention view, this paper enhances the Convolutional Block Attention Module by introducing the multi‐focal channel attention (MFCA) concept, which we call MFCA. In the proposed MFCA channel attention, focal attention spots allow us to consider scattered areas of interest more effectively. To evaluate the effectiveness of the proposed method, multiple experiments are conducted on five well‐known benchmark datasets for polyp image segmentation: Kvasir, CVC‐Clinic DB, CVC‐Colon DB, CVC‐T, and ETIS‐Larib. Various testing scenarios are examined, including the impact of two focal attention spots and four focal attention spots. The experimental results demonstrated that the proposed method achieved superior performance compared to previous state‐of‐the‐art techniques on two of the benchmark datasets, and ranked second on a third dataset. [ABSTRACT FROM AUTHOR]

Accurate landslide segmentation from remote sensing data is pivotal for efficient emergency response and risk management. In recent years, data-driven deep learning approaches have emerged as a significant area of focus in this domain. However, the limited availability of landslide data often restricts the effectiveness of these approaches. This study introduces the StyleGAN2-transformer framework for landslide segmentation, utilizing generative adversarial networks (GANs) for the first time to create synthetic, high-quality landslide images to address the data scarcity issue that undermines landslide segmentation model performance. Two datasets were developed: one containing a limited set of real landslide images and the other supplemented with synthetic landslide images generated by StyleGAN2. These datasets facilitated comparative experiments to quantitatively assess the impact of synthetic data on the performance of both convolutional neural network (CNN) and transformer series models, employing a suite of metrics for thorough evaluation. The findings indicate that adding synthetic landslide images from StyleGAN2 improves the overall accuracy of most landslide segmentation models significantly, achieving more than a 10% increase. Moreover, integrating StyleGAN2 with transformer models presents an optimized approach, as transformer models surpass CNN models in accuracy when adequate training data are available. Finally, the results also confirm that the StyleGAN2-transformer framework exhibits strong generalizability in a variety of scenarios. [ABSTRACT FROM AUTHOR]

The Bayesian optimization algorithm uses Bayesian networks as the probability model of its solution space. Although the research on this algorithm has steadily developed, there are still some problems in its application process, such as excessive computational complexity. To solve various problems in Bayesian algorithm, reduce its computational complexity, and enable it to better achieve image segmentation. The study chooses to improve the Bayesian algorithm on the basis of immune algorithm, and solves the problem of computational complexity by reducing the number of Bayesian network construction times, thereby improving the individual fitness of the population. Through simulation experiments, it has been shown that the average number of times the improved Bayesian algorithm reaches the optimal value is 30, which is higher than the traditional algorithm's 20 times. Its excellent optimization ability searches for the optimal threshold to complete image segmentation. The improved Bayesian optimization algorithm based on immune algorithm can effectively reduce computational complexity, shorten computational time, and improve convergence. And applying Bayesian algorithm to image segmentation has broadened the application field of the algorithm and found new exploration directions for image segmentation. [ABSTRACT FROM AUTHOR]

Deep learning has been widely applied in the fields of image classification and segmentation, while adversarial attacks can impact the model's results in image segmentation and classification. Especially in medical images, due to constraints from factors like shooting angles, environmental lighting, and diverse photography devices, medical images typically contain various forms of noise. In order to address the impact of these physically meaningful disturbances on existing deep learning models in the application of burn image segmentation, we simulate attack methods inspired by natural phenomena and propose an adversarial training approach specifically designed for burn image segmentation. The method is tested on our burn dataset. Through the defensive training using our approach, the segmentation accuracy of adversarial samples, initially at 54%, is elevated to 82.19%, exhibiting a 1.97% improvement compared to conventional adversarial training methods, while substantially reducing the training time. Ablation experiments validate the effectiveness of individual losses, and we assess and compare training results with different adversarial samples using various metrics. [ABSTRACT FROM AUTHOR]

Traditional thresholding methods are widely used to extract objects of interest from image backgrounds in various practical applications. However, these methods often face challenges in complex scenes due to poor uniformity, noise, and low contrast. To overcome these limitations, this paper proposes a peak-weaken Otsu method (PWOTSU) that improves the segmentation performance of the Otsu method for automatically extracting objects in complex scenes. The proposed approach uses a set of cross parameters as weights for the Otsu criterion function to adaptively weaken the between-class variance at the peak of the histogram. This ensures that an appropriate threshold value is always obtained for images with different types of histogram distribution. The improved criterion function has the advantage of obtaining a more accurate threshold value without the need for additional parameters, making it easily applicable to various practical applications. Experimental results demonstrate that the proposed method effectively improves the segmentation accuracy and robustness compared to the standard Otsu method and its modifications, as evidenced by qualitative and quantitative evaluations. [ABSTRACT FROM AUTHOR]

The non‐destructive inspection of Printed Circuit Boards (PCBs) through r‐ray Computer Tomography (CT) is a recently developed method that offers several advantages over traditional inspection techniques. This method is non‐invasive, quick, and offers high resolution, leading to significant improvements in inspection and repair efficiency. Post‐image analysis is an important step in PCB inspection and has important practical significance for automatic positioning and determining the location of faults. Usually, the results of image segmentation are an important basis for PCB defect detection, and accurate segmentation results can effectively improve the efficiency and accuracy of PCB inspection and increase the level of automation. This paper discusses two innovative improvements for the automatic segmentation process: firstly determining which slices of an x‐ray CT 3D PCB stack belong to which layer on a physical PCB in an automatic, generic and completely unsupervised way, which is verified on a 4‐layer PCB; secondly proposing a level set‐based image segmentation algorithm for the problem of gray scale inhomogeneity present in PCB CT. Experimental results on real PCB CT images with high aliasing and artifacts show that the proposed model can obtain better performance than the popular active contour models. [ABSTRACT FROM AUTHOR]

Accurately studying structural connectivity requires precise tract segmentation strategies. The U-Net network has been widely recognized for its exceptional capacity in image segmentation tasks and provides remarkable results in large tract segmentation when high-quality diffusion-weighted imaging (DWI) data are used. However, short tracts, which are associated with various neurological diseases, pose specific challenges, particularly when high-quality DWI data acquisition within clinical settings is concerned. Here, we aimed to evaluate the U-Net network ability to segment short tracts by using DWI data acquired in different experimental conditions. To this end, we conducted three types of training experiments involving 350 healthy subjects and 11 white matter tracts, including the anterior, posterior, and hippocampal commissure, fornix, and uncinated fasciculus. In the first experiment, the model was exclusively trained with high-quality data of the Human Connectome Project (HCP) dataset. The second experiment focused on images of healthy subjects acquired from a local hospital dataset, representing a typical clinical routine acquisition. In the third experiment, a hybrid training approach was employed, combining data of the HCP and local hospital datasets. Then, the best model was also tested in unseen DWIs of 10 epilepsy patients of the local hospital and 10 healthy subjects acquired on a scanner from another company. The outcomes of the third experiment demonstrated a notable enhancement in performance when contrasted with the preceding trials. Specifically, the short tracts within the local hospital dataset achieved Dice scores ranging between 0.60 and 0.65. Similar intervals were obtained with HCP data in the first experiment, and a substantial improvement compared to the scores between 0.37 and 0.50 obtained with the local hospital dataset at the same experiment. This improvement persisted when the method was applied to diverse scenarios, including different scanner acquisitions and epilepsy patients. These results indicate that combining datasets from different sources, coupled with resolution standardization strengthens the neural network ability to generalize predictions across a spectrum of datasets. Nevertheless, short tract segmentation performance is intricately linked to the training composition, to validation, and to testing data. Moreover, curved tracts have intricate structural nature, which adds complexities to their segmenting. Although the network training approach tested herein has provided promising results, caution must be taken when extrapolating its application to datasets acquired under distinct experimental conditions, even in the case of higher-quality data or analysis of long or short tracts. • To study structural connectivity, precise tract segmentation is crucial. • U-Net network is recognized for its exceptional capacity for image segmentation. • High-quality DWI data is a challenge for clinical image acquisitions. • Short tracts pose a greater challenge for neural networks to accurately predict. • Effectiveness of automatic segmentation on clinical images. [ABSTRACT FROM AUTHOR]

In response to the imbalance between normal and abnormal samples in existing anomaly detection datasets, as well as the complexity in defining anomalies, we introduce a new dataset named Remote Stop to provide data support for existing algorithms. Concurrently, we propose an unsupervised video anomaly detection method based on conditional generative adversarial networks. Our approach trains the model to learn the distribution of normal video data, enabling it to identify anomalous events. The incorporation of a spatial attention mechanism enhances the model's performance in detecting abnormal behaviors in video frames while maintaining high processing efficiency. Moreover, unlike other methods that assess the entire image, our approach uses overlapping image blocks to determine anomalies, enhancing the accuracy and robustness of the model in image segmentation. These innovations not only address the issues of scarce samples and high-cost labeling but also provide new perspectives and tools for video anomaly detection in the field of public safety. The effectiveness of the model was validated on the Avenue and Ped2 datasets and applied to our newly created dataset (Remote Stop), achieving an AUC of 84.3% and processing 61 video frames per second. This enables efficient sequential processing of large-scale video data, offering positive contributions to enhancing public road safety by providing early warnings and enabling timely preventive measures. [ABSTRACT FROM AUTHOR]

The fashion industry's traditional price-setting methods, based on historical sales and Fashion Week trends, are inadequate in the digital era. Rapid changes in collections and consumer preferences necessitate advanced Artificial Intelligence (AI) techniques. These AI methods should analyze data from various sources, including social media and e-commerce, to predict future fashion trends and prices. In this paper, we propose, apply, and assess a data analytics approach, i.e., FashionXpert, employing several image processing and machine learning techniques in an AI pipeline for garment price prediction. It integrates various heterogeneous data sources (e.g., textual and image data from e-stores, brand websites, and social media) to obtain more consistent, accurate, and beneficial information. We evaluated its effectiveness with an industrial data set obtained by a fashion search tool from the electronic commerce sites of clothing brands. FashionXpert predicted garment prices with an average Mean Absolute Error (MAE) of 15.31 EUR on a data set that has a standard deviation of 72.99 EUR. [ABSTRACT FROM AUTHOR]

In this paper, we propose a machine learning approach for detecting superficial defects in metal surfaces using point cloud data. We compare the performance of two popular deep learning architectures, multilayer perceptron networks (MLPs) and fully convolutional networks (FCNs), with varying feature sets. Our results show that FCNs (F1=0.94) outperformed MLPs (F1=0.52) in terms of precision, recall, and F1-score. We found that transfer learning with pre-trained models can improve performance when the amount of available data is limited. Our study highlights the importance of considering the amount and quality of training data in developing machine learning models for defect detection in industrial settings with 3D images. [ABSTRACT FROM AUTHOR]

Motivated by biological evolution, this paper explains the rationality of Vision Transformer by analogy with the proven practical evolutionary algorithm (EA) and derives that both have consistent mathematical formulation. Then inspired by effective EA variants, we propose a novel pyramid EATFormer backbone that only contains the proposed EA-based transformer (EAT) block, which consists of three residual parts, i.e., Multi-scale region aggregation, global and local interaction, and feed-forward network modules, to model multi-scale, interactive, and individual information separately. Moreover, we design a task-related head docked with transformer backbone to complete final information fusion more flexibly and improve a modulated deformable MSA to dynamically model irregular locations. Massive quantitative and quantitative experiments on image classification, downstream tasks, and explanatory experiments demonstrate the effectiveness and superiority of our approach over state-of-the-art methods. E.g., our Mobile (1.8 M), Tiny (6.1 M), Small (24.3 M), and Base (49.0 M) models achieve 69.4, 78.4, 83.1, and 83.9 Top-1 only trained on ImageNet-1K with naive training recipe; EATFormer-Tiny/Small/Base armed Mask-R-CNN obtain 45.4/47.4/49.0 box AP and 41.4/42.9/44.2 mask AP on COCO detection, surpassing contemporary MPViT-T, Swin-T, and Swin-S by 0.6/1.4/0.5 box AP and 0.4/1.3/0.9 mask AP separately with less FLOPs; Our EATFormer-Small/Base achieve 47.3/49.3 mIoU on ADE20K by Upernet that exceeds Swin-T/S by 2.8/1.7. Code is available at https://github.com/zhangzjn/EATFormer. [ABSTRACT FROM AUTHOR]

• Improving Coastline Edge Detail: EDS module enhances edge fitting. • Bridging Semantic Gap: CSDS and AFM collaboration enhances semantic information. • Outperforms with mIoU: CSAFNet achieves remarkable 96.72% mIoU value. The extraction of coastlines from remote sensing images is vital for promoting sustainable development in coastal areas, conserving marine environments, strengthening disaster response capabilities, and supporting scientific research. However, current coastline detection approaches using remote sensing face challenges related to resolution, terrain, boundary, and data, requiring accurate solutions for reliability. Here, we introduce the Collaborative Supervision and Attention Fusion (CSAFNet) model for pixel-level sea-land segmentation, with a primary goal of improving the accuracy of coastline extraction. The model integrates the Edge Deep Supervision (EDS) module to enhance coastline edge detail fitting. Additionally, the Collaborative Semantic Deep Supervision (CSDS) module and Attention Fusion Module (AFM) collaborate to bridge the semantic gap between different hierarchical features, resulting in a more precise and detailed delineation of coastlines. Experimental validation on the publicly available SLSD 1 1 The dataset was initially labeled as "sea-land segmentation data" by the authors. For ease of discussion, we will adopt the practice of using the initial letters of each word as an abbreviation. dataset has demonstrated superiority over various advanced methods, with an impressive mIoU value of 96.72%. Through simple optimization, detailed and rich coastlines can be extracted, validating the feasibility of coastline extraction. [ABSTRACT FROM AUTHOR]

Manual delineation of organs at risk and clinical target volumes is essential in radiotherapy planning. Atlas-based auto-segmentation (ABAS) algorithms have become available and been shown to provide accurate contouring for various anatomical sites. Recently, deep learning auto-segmentation (DL-AS) algorithms have emerged as the state-of-the-art in medical image segmentation. This study aimed to evaluate the effect of auto-segmentation on the clinical workflow for contouring different anatomical sites of cancer, such as head and neck (H&N), breast, abdominal region, and prostate. Patients with H&N, breast, abdominal, and prostate cancer (n = 30 each) were enrolled in the study. Twenty-seven different organs at four sites were evaluated. RayStation was used to apply the ABAS. Siemens AI-Rad Companion Organs RT was used to apply the DL-AS. Evaluations were performed with similarity indices using geometric methods, time-evaluation, and qualitative scoring visual evaluations by radiation oncologists. The DL-AS algorithm was more accurate than ABAS algorithm on geometric indices for half of the structures. The qualitative scoring results of the two algorithms were significantly different, and DL-AS was more accurate on many contours. DL-AS had 41%, 29%, 86%, and 15% shorter edit times in the HnN, breast, abdomen, and prostate groups, respectively, than ABAS. There were no correlations between the geometric indices and visual assessments. The time required to edit the contours was considerably shorter for DL-AS than for ABAS. Auto-segmentation with deep learning could be the first step for clinical workflow optimization in radiotherapy. [ABSTRACT FROM AUTHOR]

Purpose: Central venous catheterization (CVC) carries inherent risks which can be mitigated through the use of appropriate ultrasound-guidance during needle insertion. This study aims to comprehensively understand patient anatomy as it is visualized during CVC by employing a semi-automated image analysis method to track the internal jugular vein and carotid artery throughout recorded ultrasound videos. Methods: The ultrasound visualization of 50 CVC procedures were recorded at Penn State Health Milton S. Hershey Medical Center. The developed algorithm was used to detect the vessel edges, calculating metrics such as area, position, and eccentricity. Results: Results show typical anatomical variations of the vein and artery, with the artery being more circular and posterior to the vein in most cases. Notably, two cases revealed atypical artery positions, emphasizing the algorithm's precision in detecting anomalies. Additionally, dynamic vessel properties were analyzed, with the vein compressing on average to 13.4% of its original size and the artery expanding by 13.2%. Conclusion: This study provides valuable insights which can be used to increase the accuracy of training simulations, thus enhancing medical education and procedural expertise. Furthermore, the novel approach of employing automated data analysis techniques to clinical recordings showcases the potential for continual assessment of patient anatomy, which could be useful in future advancements. [ABSTRACT FROM AUTHOR]

In medical clinical scenarios for reasons such as patient privacy, information protection and data migration, when domain adaptation is needed for real scenarios, the source-domain data is often inaccessible and only the pre-trained source model on the source-domain is available. Existing solutions for this type of problem tend to forget the rich task experience previously learned on the source domain after adapting, which means that the model simply overfits the target-domain data when adapting and does not learn robust features that facilitate real task decisions. We address this problem by exploring the particular application of source-free domain adaptation in medical image segmentation and propose a two-stage additive source-free adaptation framework. We generalize the domain-invariant features by constraining the core pathological structure and semantic consistency between different perspectives. And we reduce the segmentation generated by locating and filtering elements that may have errors through Monte-Carlo uncertainty estimation. We conduct comparison experiments with some other methods on a cross-device polyp segmentation and a cross-modal brain tumor segmentation dataset, the results in both the target and source domains verify that the proposed method can effectively solve the domain offset problem and the model retains its dominance on the source domain after learning new knowledge of the target domain.This work provides valuable exploration for achieving additive learning on the target and source domains in the absence of source data and offers new ideas and methods for adaptation research in the field of medical image segmentation. [ABSTRACT FROM AUTHOR]

Segmenting autophagic bodies in yeast TEM images is a key technique for measuring changes in autophagosome size and number in order to better understand macroautophagy/autophagy. Manual segmentation of these images can be very time consuming, particularly because hundreds of images are needed for accurate measurements. Here we describe a validated Cellpose 2.0 model that can segment these images with accuracy comparable to that of human experts. This model can be used for fully automated segmentation, eliminating the need for manual body outlining, or for model-assisted segmentation, which allows human oversight but is still five times as fast as the current manual method. The model is specific to segmentation of autophagic bodies in yeast TEM images, but researchers working in other systems can use a similar process to generate their own Cellpose 2.0 models to attempt automated segmentations. Our model and instructions for its use are presented here for the autophagy community. Abbreviations: AB, autophagic body; AvP, average precision; GUI, graphical user interface; IoU, intersection over union; MVB, multivesicular body; ROI, region of interest; TEM, transmission electron microscopy; WT,wild type. [ABSTRACT FROM AUTHOR]

Semantic nuclei segmentation is a challenging area of computer vision. Accurate nuclei segmentation can help medics in diagnosing many diseases. Automatic nuclei segmentation can help medics in diagnosing many diseases such as cancer by providing automatic tissue analysis. Deep learning algorithms allow automatic feature extraction from medical images, however, hematoxylin and eosin (H&E) stained images are challenging due to variability in staining and textures. Using pre-trained models in deep learning speeds up development and improves their performance. This paper compares Deeplabv3+ and U-Net deep learning methods with the pre-trained models ResNet-50 and EfficientNetB4 embedded in their architecture. In addition, different regularization and dropout parameters are applied to prevent overtraining. The experiment was conducted on the PanNuke dataset consisting of nearly 8,000 histological images and annotated nuclei. As a result, the ResNet50-based DeepLabV3+ model with L2 regularization of 0.02 and dropout of 0.7 showed efficiency with dice coefficient (DCS) of 0.8356, intersection over union (IOU) of 0.7280, and loss of 0.3212 on the test set. [ABSTRACT FROM AUTHOR]

Medical imaging plays a crucial role in diagnosing patient conditions, with magnetic resonance imaging (MRI) standing as a significant modality for numerous years. However, leveraging convolutional neural network (CNN) architectures like U-Net and its variations for anatomical segmentation demands considerable memory, particularly when working with full 3D image sets. Therefore, downsampling 3D MRIs proves advantageous in reducing memory consumption. Nevertheless, downsampling leads to a reduction in voxel count, potentially impacting the performance of commonly used segmentation metrics. The jaccard similarity index (JSI), dice similarity coefficient (DSC), and structural similarity index (SSIM) are extensively employed in image segmentation contexts. Hence, this study employs all three metrics to assess downsampled images and evaluate the robustness of the metrics when used to evaluate the downsampled 3D MRI images. The results show that JSI and DSC are more robust than SSIM when handling the downsampled data. [ABSTRACT FROM AUTHOR]

The development of deep learning technology is widely used for various purposes, including recognizing characters in a document. One of the scripts that can benefit from this deep learning technology is the Komering script, which is a local script in the South Sumatra region. However, there are challenges in reading documents written in this script, requiring a method to separate each character in a document. Therefore, there is a need for a technology that can automatically segment images of documents written in the Komering script. This research introduces an innovative technique for segmenting images of characters in documents that contain Komering script characters. The segmentation technique employs bounding box technology to separate each Komering script character, subsequently recognized by a pre-trained deep learning model. The bounding box approach imposes restrictions on the segmented object area. To recognize Komering characters, a deep learning model with a convolutional neural network (CNN) algorithm is employed. [ABSTRACT FROM AUTHOR]

Divergence measure between intuitionistic fuzzy sets (IFSs) is important due to its wide range of applications in various fields including pattern recognition, image segmentation, decision-making and clustering. This paper introduces the characterization of belief and plausible intuitionistic fuzzy sets (BP-IFSs) to explore novel logarithmic and non-logarithmic divergence measures between two BP-IFSs. These measures are regarded as highly useful approaches to express ambiguous information within the framework of Dempster–Shafer Theory (DST). An axiomatic definition based on proposed divergence measures is also stated within a frame work of newly established theory. Furthermore, the proposed divergence measures are utilized in three different applications: (i) an example related to the recognition of BP-IFS patterns is provided to demonstrate the practicality of the proposed method in pattern recognition. (ii) An example of Hierarchical agglomerative clustering is also provided. (iii) Introduces an innovative Belief and Plausible Combinative Distance-based Assessment (BP-CODAS) method based on proposed measures for resolving Multicriteria Decision Making (MCDM) problems connected to child labor in under developed countries. The examples provided in these different directions are sufficient to demonstrate the effectiveness, applicability and viability of the suggested methods within the framework of generalized DST. [ABSTRACT FROM AUTHOR]

The recognition of telugu hand written text is been one of the problems in many applications. To overcome the problem a deep learning technique is proposed in this work i.e. a Dense convolutional neural network (DCNN) model. A telugu dataset which is taken form IIIT-HW-Telugu is utilized to perform the proposed model. In this paper a four stage telugu word retrieval is performed, initially thinning of image is performed using morphological operation, secondly Densenet-CNN is applied for thinning image, thirdly perform OCR based image segmentation, finally two models like HARRIS and BRISK features to extract the features and evaluate information from the given input HWT images. The parameters evaluated are hamming distance, PSNR, MSE, Noise Sensitivity and rate of thinning. The proposed model outperformed well compared to other methods. The PSNR obtained using proposed model is 54.74, hamming distance is 1.2. [ABSTRACT FROM AUTHOR]

Diagnosing and segmenting brain tumors manually through MRI imaging is a complex and time-consuming process. However, advancements in machine learning (ML) and deep learning (DL) technologies have enabled the automatic identification and categorization of brain tumors using computer-aided design. This study utilizes MRI data to develop a system for the automatic identification and categorization of brain tumors based on region-based convolutional neural networks (R-CNN). The proposed R-CNN approach, coupled with Chicken Swarm Optimization (CSO) technique, enables the identification and classification of brain tumors into stages. This method involves processing, segmenting, extracting features, and organizing the MRI images. Image preparation includes adaptive fuzzy filtering (AFF) to eliminate noise and enhance the quality of MRI images. To detect regions of brain injury, MRI scans undergo cranial segmentation and classification (CSO) based on Tsallis entropy-based image segmentation. A Residual Network (ResNet) is employed to fuse handcrafted and deep features, generating a meaningful set of feature vectors. Extensive simulations are conducted on the BRATS 2015 dataset to evaluate the improved performance in classifying brain tumors. The RCNN-CSO method demonstrates superior performance compared to other contemporary techniques, achieving a precision of 92.35%, sensitivity of 93.52%, specificity of 94.52 % and an accuracy of 96 %. This represents a significant improvement in brain tumor classification and its outomst accuracy. [ABSTRACT FROM AUTHOR]

Cracks are a common problem in concrete surfaces. With the continuous optimization of machine vision-based inspection systems, effective crack detection and recognition is the core of the entire system. In this study, support vector machine (SVM) was used to distinguish cracks from other regions. To complete the recognition system of the SVM, a framework consisting of an image processing and recognition model was proposed. An algorithm combining the Prewitt operator with the Otsu threshold was proposed for image segmentation. The binary image processed by the new algorithm combined with mathematical morphology can result in a more complete crack zone and fewer interference regions. After the initial parameter extraction, most of the impurity areas were screened by preliminary discrimination, removing 99% of the impurities. This processing step ensured the balance and effectiveness of the samples. To establish an automatic identification model based on SVM with a radial basis function, compactness, occupancy rate, and length–width ratio were selected as input parameters after comparing these three features with all six features of the crack. The recognition accuracy of this system reaches 97.14%, demonstrating that the proposed method is effective and satisfies practical requirements. [ABSTRACT FROM AUTHOR]