Segment Boundaries Research Articles

TransPV: Refining photovoltaic panel detection accuracy through a vision transformer-based deep learning model

The increasing need to develop renewable energy sources to combat climate change has led to a significant rise in demand for photovoltaic (PV) installations. Consequently, accurately detecting and estimating the capacity and potential for electricity generation of these installed PV systems has become crucial for effective energy management. However, due to the limitation of localized mechanism in convolution operations, traditional FCN-based methods face challenges in capturing global dependencies among remote sensing image patches. As a result, they struggle to model long-range interactions between different PV panels with diverse structures, including varying size, shape, and texture, particularly in distributed residential areas. To tackle the challenge of modeling PV panels with diverse structures, we propose a coupled U-Net and Vision Transformer model named TransPV for refining PV semantic segmentation. Specifically, Mix Transformer block is incorporated in the encoder to enhance the modeling of global context, while the U-Shaped structure enables the combination of multi-level features, resulting in enriched feature representation, which helps the model better understand objects in the images, regardless of their shape, size, and texture. In addition, we implement the PointRend module in the decoder to obtain finer segmentation boundary details, and utilize a novel refiner loss function during the training process to alleviate the problem of extremely unbalanced samples. The experimental results from the Heilbronn datasets demonstrate the remarkable performance of our proposed TransPV model in addressing intra-class diversity of PV structures, surpassing previous state-of-the-art methods with an impressive IoU and accuracy of 0.802 and 0.876, respectively. Furthermore, our model exhibits high generalization capability on the BDPV dataset with IoU of 0.745. The obtained results highlight the superiority of TransPV in improving accuracy and addressing diverse structure issues in PV segmentation, which provides valuable insights for optimizing the efficiency and sustainability of renewable energy.

Knowledge-based U-Net and transfer learning for automatic boundary segmentation

An effective soil classification method is essential for soil layering interpretation. Traditional methods rely on empirical formulae or a general unified system, which may not be accurate for every site. Recent advancements in deep learning have shown promising results in a wide range of practical domains, which provides motivation to explore its potential application in soil classification. A “U-shaped” convolutional neural network (U-Net), coupled with multiple source data to incorporate prior knowledge, is adopted in this study to predict the depths of soil stratum boundaries based on cone penetration testing (CPT), standard penetration testing (SPT) and laboratory index testing data. The U-Net model is first pre-trained on open-access data from other sites around the world and then trained through transfer learning on datasets gathered specifically for the Suzhou No.6 metroline project. Comparison is made from different metrics to investigate how pre-training benefits model performance. The soil boundaries in nine selected output images are extracted and reversed to their original depth for visualization of a subsurface profile. Comparisons to the benchmark interpretations show that the predicted soil profiles have reasonably good agreement with the benchmark profiles developed by an engineering expert. The results indicate that the proposed method is effective and efficient for the prediction of soil stratification.

EPIC: An epidemiological investigation of COVID-19 dataset for Chinese named entity recognition

Since the outbreak of COVID-19, it has had a huge impact on the whole world. In China, there have been a large number of epidemiological investigation reports in response to COVID-19. In order to more effectively analyze and utilize these reports for future large-scale epidemics, this paper proposes a new three-stage annotation method and utilizes the method to construct an epidemiological investigation of COVID-19 dataset for Chinese named entity recognition (CNER)—EPIC (EPidemiological Investigation of COVID-19). EPIC contains 10 categories of named entities, focusing on the travel history of confirmed cases. The corpus of the EPIC consists of 226 official epidemiological investigation reports and the inter-annotator agreement of the EPIC reaches 0.97. Based on EPIC, this paper proposes the Semantic Embedding with Chinese Character Structural Features (SECCSF) method to improve the accuracy of segmentation boundary detection and entity category determination in CNER. In the experimental phase, this paper implements several baselines to conduct experiments on EPIC. The baseline with the SECCSF method achieves an F1 value of 0.892. This indicates the effectiveness of the SECCSF method for the NER task on the EPIC. We release the EPIC at: https://github.com/tinyyhorm/EPIC.

A cell image segmentation method based on edge feature residual fusion.

In recent years, cancer has seriously damaged human health, and the morphological structure of cells serves as an important basis for cancer diagnosis and grading. Automatic cell segmentation based on deep learning has become an important means of computer-aided pathological diagnosis. Aiming at the existing problems of rough segmentation boundaries and inaccurate segmentation in cell image segmentation, this paper designs a cell image segmentation network model (ERF-TransUNet) based on edge feature residual fusion from the perspective of mutual complementarity and constraint between edge features and object features. The model uses a hybrid architecture of CNN and Transformer to extract multi-scale features from cell images, and adds independent edge feature extraction modules and residual fusion modules to enhance the extraction of edge features and their constraints when fusing with cell object features, improving the accuracy of cell contour positioning. Through experiments on two gland cell datasets, CRAG and Glas, and comparing the segmentation effects with current popular deep learning models, the network model proposed in this paper has achieved good performance in both Dice coefficient and Hausdorff distance, which can effectively improve the segmentation effect of cell images.

Photometry, Centroid and Point-spread Function Measurements in the LSST Camera Focal Plane Using Artificial Stars

The Vera C. Rubin Observatory’s LSST Camera (LSSTCam) pixel response has been characterized using laboratory measurements with a grid of artificial stars. We quantify the contributions to photometry, centroid, point-spread function size, and shape measurement errors due to small anomalies in the LSSTCam CCDs. The main sources of those anomalies are quantum efficiency variations and pixel area variations induced by the amplifier segmentation boundaries and “tree-rings”—circular variations in silicon doping concentration. This laboratory study using artificial stars projected on the sensors shows overall small effects. The residual effects on point-spread function (PSF) size and shape are below 0.1%, meeting the ten-year LSST survey science requirements. However, the CCD mid-line presents distortions that can have a moderate impact on PSF measurements. This feature can be avoided by masking the affected regions. Effects of tree-rings are observed on centroids and PSFs of the artificial stars and the nature of the effect is confirmed by a study of the flat-field response. Nevertheless, further studies of the full-focal plane with stellar data should more completely probe variations and might reveal new features, e.g., wavelength-dependent effects. The results of this study can be used as a guide for the on-sky operation of LSSTCam.

Direct imposition of the wall boundary condition for weakly compressible flows in three-dimensional smoothed particle hydrodynamics simulations

This study introduces a novel method for imposing wall boundary conditions in smoothed particle hydrodynamics (SPH). SPH is a particle method based on the Lagrangian approach, primarily employed in fluid analysis as a part of numerical computation methods. Due to its ability to discretize space using particles, SPH excels in handling analyses of free surface flow or multiphase flow with intricate boundary surfaces. However, there is a drawback in modeling wall boundaries using particles, as resolving the particle deficiency problem necessitates multi-layered boundary particles to be arranged behind the wall boundary. This leads to difficulties in implementing complex shapes and adds computational expense. To address this issue, this study suggests the use of boundary segments for wall boundary modeling and specifically employs triangular segments for three-dimensional expansions. For robust application of boundary conditions, a method considering both Poisson's equation and geometric configurations is proposed. The proposed method is independent of the segment density, which facilitates efficient and flexible modeling. In addition, by imposing accurate boundary conditions from the wall, the stability and accuracy of the solution are enhanced. The performance of the proposed method is validated through numerical examples, compared with various analytical and experimental results.

Deformation behaviour, microstructure evolution and dynamic recrystallization mechanism of AZ31 magnesium alloy under co-extruded by Mg-Al composite billet

Conventional methods for hot extrusion of magnesium (Mg) alloy sheets have limitations for microstructure improvement and mechanical property enhancement. Co-extrusion of Mg alloy with an Al alloy core provides an opportunity for refining the grain size and improving basal texture. This study introduces a novel severe plastic deformation (SPD) method, Direct Extrusion-Continuous Shear (DE-CS), utilizing 6063 Al alloy as the core to fabricate innovative Mg/Al clad composite sheets. Monolithic Mg sheets (non-co-extruded condition) are extruded under identical conditions for comparison. By evaluating the microstructure and mechanical properties of AZ31 Mg alloy under co-extrusion and non-co-extrusion conditions, the positive effect of Mg-Al composite billet co-extrusion on AZ31 Mg alloy is established. The grain refinement effect in co-extruded samples is superior, enhancing plasticity without significant reduction in tensile strength. The microstructure evolution and dynamic recrystallization (DRX) behaviour of AZ31 Mg alloy during co-extrusion were analyzed comprehensively. The results indicate that, in the initial stages of deformation, various variants of {10−12} extension twins (ETs) play a crucial role in refining the original grains, forming initial [10−10]//ED fiber texture components alongside other non-basal components. Upon entering the shear deformation (SD) zone, deformed twins disappear, and small dynamically recrystallized (DRXed) grains emerge along boundaries of coarse deformation grains and within the grains. Strong shear deformations and compressive stress from transverse direction (TD) induce intensified prismatic <a> and pyramidal <c+a> slip, promoting continuous dynamic recrystallization (CDRX) and discontinuous dynamic recrystallization (DDRX). CDRX predominantly affects elongated deformed grains, forming an [10−10]//ED fiber texture through high-energy grain boundary segmentation. Subsequently, compatible stress increases due to the activity of non-basal slip dislocation and the reduction in grain boundary (GB) size. DDRX emerges as the dominant mechanism, generating randomized [10−10]//ED fiber and non-fiber textures within parent grains and at boundaries.

A cutting width measurement method for the unmanned rice harvester based on RGB-D images

Currently, the unmanned rice harvester based on satellite navigation is affected by tracked chassis steering errors and navigation accuracy, which can result in harvesting omissions and undercutting. Therefore, we proposed a cutting width measurement method for unmanned rice harvesters based on RGB-D images. The proposed method can provide navigational aids for driving and may also provide a basis for feed quantity detection and adjustment. We designed a lightweight unharvested area segmentation model (UANet) with a mean pixel accuracy (mPA) and a mean intersection over union (mIoU) of 98.28% and 97.15%, respectively, which are both higher than the lightweight semantic segmentation models, such as DABNet, CFPNet-V2 and ENet. We implemented unharvested area segmentation and harvesting boundary points extraction on RGB images, combined with depth maps to convert the pixel coordinates of harvesting boundary points into 3D world coordinates for cutting width measurement. Through experiments on rice harvest field images under different working conditions, the results showed that the success rate of harvesting boundary points detection was 98.33%, the error of cutting width measurement was within 2.34%, and the speed of the algorithm was approximately 9.7 fps. The proposed method can be adapted to complex harvesting scenarios and has the advantages of low hardware cost, high measurement accuracy, and good robustness, which can be extended to wheat.

Multi-scale context UNet-like network with redesigned skip connections for medical image segmentation

Background and ObjectiveMedical image segmentation has garnered significant research attention in the neural network community as a fundamental requirement for developing intelligent medical assistant systems. A series of UNet-like networks with an encoder-decoder architecture have achieved remarkable success in medical image segmentation. Among these networks, UNet2+ (UNet++) and UNet3+ (UNet+++) have introduced redesigned skip connections, dense skip connections, and full-scale skip connections, respectively, surpassing the performance of the original UNet. However, UNet2+ lacks comprehensive information obtained from the entire scale, which hampers its ability to learn organ placement and boundaries. Similarly, due to the limited number of neurons in its structure, UNet3+ fails to effectively segment small objects when trained with a small number of samples. MethodIn this study, we propose UNet_sharp (UNet#), a novel network topology named after the “#” symbol, which combines dense skip connections and full-scale skip connections. In the decoder sub-network, UNet# can effectively integrate feature maps of different scales and capture fine-grained features and coarse-grained semantics from the entire scale. This approach enhances the understanding of organ and lesion positions and enables accurate boundary segmentation. We employ deep supervision for model pruning to accelerate testing and enable mobile device deployment. Additionally, we construct two classification-guided modules to reduce false positives and improve segmentation accuracy. ResultsCompared to current UNet-like networks, our proposed method achieves the highest Intersection over Union (IoU) values ((92.67±0.96)%, (92.38±1.29)%, (95.36±1.22)%, (74.01±2.03)%) and F1 scores ((91.64±1.86)%, (95.70±2.16)%, (97.34±2.76)%, (84.77±2.65)%) on the semantic segmentation tasks of nuclei, brain tumors, liver, and lung nodules, respectively. ConclusionsThe experimental results demonstrate that the reconstructed skip connections in UNet successfully incorporate multi-scale contextual semantic information. Compared to most state-of-the-art medical image segmentation models, our proposed method more accurately locates organs and lesions and precisely segments boundaries.

CHBS-Net: 3D Point Cloud Segmentation Network with Key Feature Guidance for Circular Hole Boundaries

In laser scanning inspection systems for sheet metal parts, the rapid and accurate inspection of the high-precision holes is not only crucial but difficult. The accuracy of the circular holes, especially the locating datum holes on the parts, plays an important role in the assembly quality. However, accurately segmenting the circular hole boundary points required for circular hole fitting from large-scale scanning point cloud data remains one of the most difficult tasks for inspection accuracy improvement. To address this problem, a segmentation network called the circular hole boundary segmentation network (CHBS-Net) is proposed for boundary point cloud extraction. Firstly, an encoding–decoding–attention (EDA) fusion guidance mechanism is used to address the imbalance in data distribution due to the small proportion of boundary points in the overall point cloud. Secondly, a long short-term memory (LSTM) network parallel structure is used to capture the contour continuity and temporal relationships among boundary points. Finally, the interference of neighboring points and noise is reduced by extracting features in the multi-scale neighborhood. Experiments were performed using real cases from a sheet metal parts dataset to illustrate the procedures. The results showed that the proposed method achieves better performance than the benchmark state-of-the-art methods. The circular hole inspection accuracy is effectively improved by enhancing the segmentation accuracy of the scanning boundary points.

The peeling behavior of a heterogeneous elastic film on a rigid substrate

Although the interfacial peeling behavior of a homogeneous film-substrate system is well understood, the peeling process of a heterogeneous elastic film bonded to a rigid substrate is still unclear. In the present paper, the peeling behavior of a heterogeneous film adhering on a rigid substrate under a vertical peeling force is investigated theoretically, in which the film’s heterogeneity is characterized by the change of bending stiffness along the length of the film. Based on the principle of minimum potential energy, the typical relationship between the peeling force and the peeling displacement during the whole peeling process is achieved. Different from the homogeneous film case, the peeling force can be regulated dramatically by the film’s heterogeneous bending stiffness during the peeling process. When the interfacial debonding front propagates from a compliant segment to a stiff segment, the peeling force is enhanced; while the peeling force is weakened as the interfacial debonding front goes from a stiff segment to a compliant one. The enhancement or weakening of the peeling force depends not only on the bending stiffness of each segment of the film but also on the bending stiffness ratio of neighboring segments, as well as the adhesion length of each segment and the interfacial adhesion properties. The mechanism underlying the heterogeneity-induced tunable peeling force is further elucidated by the change rate of elastic bending energy stored in the film and energy overcoming the interfacial interaction potential during the peeling process. The transfer and redistribution of the elastic bending energy are the main factors that induce the change of peeling force as the interfacial debonding front reaches the boundary of neighboring segments with different bending stiffness. The results of this paper can provide a new strategy for designing film-substrate systems with adjustable adhesion without changing the interfacial characteristics.

An interactive cortical architecture for perceptual organization by accentuation

Accentuation has been proposed as a general principle of perceptual organization. Here, we have developed a neurodynamic architecture to explain how accentuation affects boundary segmentation and shape perception. The model consists of bottom-up and top-down pathways. Bottom-up processing involves a set of feature maps that compute bottom-up salience of surfaces, boundaries, boundary completions, and junctions. Then, a feature-based winner-take-all network selects the most salient locations. Top-down processing includes an object-based attention stage that allows enhanced neural activity to propagate from the most salient locations to all connected locations, and a visual segmentation stage that employs inhibitory connections to segregate boundaries into distinct maps. The model was tested on a series of computer simulations showing how the position of the accent affects boundary segregation in the square-diamond and the pointing illusion. The model was also tested on a variety of texture segregation tasks, showing that its performance was comparable to that of human observers. The model suggests that there is an intermediate stage of visual processing between perceptual grouping and object recognition that helps the visual system choose between competing percepts of the ambiguous stimulus.

Development of a Three-Dimensional Carotid Ultrasound Image Segmentation Workflow for Improved Efficiency, Reproducibility and Accuracy in Measuring Vessel Wall and Plaque Volume and Thickness.

Automated segmentation of carotid lumen-intima boundary (LIB) and media-adventitia boundary (MAB) by deep convolutional neural networks (CNN) from three-dimensional ultrasound (3DUS) images has made assessment and monitoring of carotid atherosclerosis more efficient than manual segmentation. However, training of CNN still requires manual segmentation of LIB and MAB. Therefore, there is a need to improve the efficiency of manual segmentation and develop strategies to improve segmentation accuracy by the CNN for serial monitoring of carotid atherosclerosis. One strategy to reduce segmentation time is to increase the interslice distance (ISD) between segmented axial slices of a 3DUS image while maintaining the segmentation reliability. We, for the first time, investigated the effect of ISD on the reproducibility of MAB and LIB segmentations. The intra-observer reproducibility of LIB and MAB segmentations at ISDs of 1 mm and 2 mm was not statistically significantly different, whereas the reproducibility at ISD = 3 mm was statistically lower. Therefore, we conclude that segmentation with an ISD of 2 mm provides sufficient reliability for CNN training. We further proposed training the CNN by the baseline images of the entire cohort of patients for automatic segmentation of the follow-up images acquired for the same cohort. We validated that segmentation with this time-based partitioning approach is more accurate than that produced by patient-based partitioning, especially at the carotid bifurcation. This study forms the basis for an efficient, reproducible, and accurate 3DUS workflow for serial monitoring of carotid atherosclerosis useful in risk stratification of cardiovascular events and in evaluating the efficacy of new treatments.

Results of a survey of ophthalmologists on the special aspects in performing and interpreting optical coherence tomography examination of the macula

BACKGROUND: Optical coherence tomography (OCT) is the main method for in vivo examination of the morphology of ocular posterior segment structures, used in modern ophthalmological practice.&#x0D; AIM: Study of special aspects in performing and interpreting the results of OCT of the macula by ophthalmologists working in the healthcare system of the Russian Federation.&#x0D; MATERIALS AND METHODS: In the anonymous survey, conducted using an original online questionnaire of 12 questions 59 respondents took part. They confirmed being ophthalmologists with experience in performing OCT and gave their consent to took part in the survey.&#x0D; RESULTS: The majority of respondents (79.6%) reported that they face the need to perform OCT without obtaining the patients clinical data. 66.1% of respondents always evaluate the correct placement of segmentation boundaries. Only 56% of the respondents always or often perform frontal (en face) scans during OCT examination of the macula. At the same time, 16.9% of specialists indicated that they did not know what segmentation boundaries are, and 8.5% noted that they did not understand the term en face scan. The principle of layer-by-layer analysis is used by 94.9% of respondents when describing images. Significant discrepancies were also revealed in the interpretation by the respondents of some parameters on specific examples of OCT scans. 71.2% of respondents have experience in using templates when describing OCT images, and 13.6% use them constantly. The proportion of specialists who consider it expedient to develop and introduce into clinical practice a standard protocol (template) for describing macular OCT images was 84.7%.&#x0D; CONCLUSIONS: Insufficient awareness of some ophthalmologists about particular principles of interpreting the results of an OCT imaging and about certain practical aspects of its application was revealed. This indicates the need in improving the quality of informational support and training of specialists. A request was received from ophthalmologists about the need to develop a standard protocol (template) for describing OCT images of the macular area with the aim of subsequent implementation into the clinical practice of the healthcare system of the Russian Federation.

Generative adversarial networks assisted machine learning based automated quantification of grain size from scanning electron microscope back scatter images

In this research, a novel approach for the automated quantification of grain size in single-phase materials using scanning electron microscopy (SEM) based back scattering electron (BSE) images is presented. Grain size analysis is crucial in understanding material properties, but manual analysis is time-consuming and subject to human error. The proposed method leverages Generative Adversarial Networks (GANs) to automatically segment grains to enable grain size quantification thereby substantially enhancing the efficiency and accuracy of the process. The grain boundary segmentation is achieved by treating the problem of grain segmentation as an ‘image translation task’ in which GANs are known to be efficient. Due to the learned ‘perceptual ability’ of the trained network, the model was able to delineate the grain boundaries even from low contrast regions and regions obscured due to sample preparation artefacts, enabling better segmentation performance over the conventional algorithms. Application of the present approach to test data set has shown mean grain size estimation accuracy of 92% and grain size distribution similarity of 80% against EBSD (Electron Back Scatter Diffraction) ground truth data. The ability of the approach in enabling reliable quantification of the microstructures even from noisy images has been demonstrated.

A Robust Deep Learning Approach for Accurate Segmentation of Cytoplasm and Nucleus in Noisy Pap Smear Images

Cervical cancer poses a significant global health burden, affecting women worldwide. Timely and accurate detection is crucial for effective treatment and improved patient outcomes. The Pap smear test has long been a standard cytology screening method, enabling early cancer diagnosis. However, to enhance quantitative analysis and refine diagnostic capabilities, precise segmentation of the cervical cytoplasm and nucleus using deep learning techniques holds immense promise. This research focuses on addressing the primary challenge of achieving accurate segmentation in the presence of noisy data commonly encountered in Pap smear images. Poisson noise, a prevalent type of noise, corrupts these images, impairing the precise delineation of the cytoplasm and nucleus. Consequently, segmentation boundaries become indistinct, leading to compromised overall accuracy. To overcome these limitations, the utilization of U-Net, a deep learning architecture specifically designed for automatic segmentation, has been proposed. This approach aims to mitigate the adverse effects of Poisson noise on the digitized Pap smear slides. The evaluation of the proposed methodology involved a dataset of 110 Pap smear slides. The experimental results demonstrate that the proposed approach successfully achieves precise segmentation of the nucleus and cytoplasm in noise-free images. By preserving the boundaries of both cellular components, the method facilitates accurate feature extraction, thus contributing to improved diagnostic capabilities. Comparative analysis between noisy and noise-free images reveals the superiority of the presented approach in terms of segmentation accuracy, as measured by various metrics, including the Dice coefficient, specificity, sensitivity, and intersection over union (IoU). The findings of this study underline the potential of deep-learning-based segmentation techniques to enhance cervical cancer diagnosis and pave the way for improved quantitative analysis in this critical field of women’s health.

CatStep: Automated Cataract Surgical Phase Classification and Boundary Segmentation Leveraging Inflated 3D-Convolutional Neural Network Architectures and BigCat

Accurate identification of surgical phases during cataract surgery is essential for improving surgical feedback and performance analysis. Time spent in each surgical phase is an indicator of performance, and segmenting out specific phases for further analysis can simplify providing both qualitative and quantitative feedback on surgical maneuvers. Retrospective surgical video analysis. One hundred ninety cataract surgical videos from the BigCat dataset (comprising nearly 4 million frames, each labeled with 1 of 11 nonoverlapping surgical phases). Four machine learning architectures were developed for segmentation of surgical phases. Models were trained using cataract surgical videos from the BigCat dataset. Models were evaluated using metrics applied to frame-by-frame output and, uniquely in this work, metrics applied to phase output. The final model, CatStep, a combination of a temporally sensitive model (Inflated 3D Densenet) and a spatially sensitive model (Densenet169), achieved an F1-score of 0.91 and area under the receiver operating characteristic curve of 0.95. Phase-level metrics showed considerable boundary segmentation performance with a median absolute error of phase start and end time of just 0.3 seconds and 0.1 seconds, respectively, a segmental F1-score @70 of 0.94, an oversegmentation score of 0.89, and a segmental edit score of 0.92. This study demonstrates the feasibility of high-performance automated surgical phase identification for cataract surgery and highlights the potential for improved surgical feedback and performance analysis. Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Design of a Novel Skin Lesion Predictor Model Using Hybrid Particle Swarm Optimization and Convolutional Neural Networks

The prediction of skin cancer at the earlier stage is extremely essential for melanoma. There is a need for intellectual computer analysis for skin lesions. The segmentation of lesion boundaries is vital to accurately identify the lesions from the dermoscopic images where the diagnosis is complex for various skin lesion types. Thus, some pre-processing steps are required to attain higher sensitive lesion boundary segmentation and classification. Initially, pre-processing is done with median filter to offer reputation of boundary preservation and does not in-cooperate newer pixel values to the processed image. Next is the process of image segmentation for Region of Interest (ROI) and non-Region of Interest ([Formula: see text] using the Jaccard Distance segmentation process. The novelty of the work relies on the inclusion of compression techniques to make the access easier without any loss. The extracted regions are encoded with Freeman Chain Coding and compressed with Lempel–Ziv–Welch (LZW) and Zero Tree Wavelet (EZW) for ROI and non-Region of Interest regions. Finally, image classification is done with Hybrid Particle Swarm Optimization and Convolutional Neural Networks ([Formula: see text]PSO-CNN). The simulation is done with TensorFlow &amp; Python environment and the proposed model outperforms the existing standard approaches. Some metrics like objective function, confusion matrix, accuracy, precision, [Formula: see text]-measure, and recall are evaluated. The model attains 77.5% accuracy, 86.36% precision, 77.5% recall and 77.92% [Formula: see text]-measure for proposed [Formula: see text]PSO-CNN which is higher than the standard CNN model.

DMCA-GAN: Dual Multilevel Constrained Attention GAN for MRI-Based Hippocampus Segmentation.

Precise segmentation of the hippocampus is essential for various human brain activity and neurological disorder studies. To overcome the small size of the hippocampus and the low contrast of MR images, a dual multilevel constrained attention GAN for MRI-based hippocampus segmentation is proposed in this paper, which is used to provide a relatively effective balance between suppressing noise interference and enhancing feature learning. First, we design the dual-GAN backbone to effectively compensate for the spatial information damage caused by multiple pooling operations in the feature generation stage. Specifically, dual-GAN performs joint adversarial learning on the multiscale feature maps at the end of the generator, which yields an average Dice coefficient (DSC) gain of 5.95% over the baseline. Next, to suppress MRI high-frequency noise interference, a multilayer information constraint unit is introduced before feature decoding, which improves the sensitivity of the decoder to forecast features by 5.39% and effectively alleviates the network overfitting problem. Then, to refine the boundary segmentation effects, we construct a multiscale feature attention restraint mechanism, which forces the network to concentrate more on effective multiscale details, thus improving the robustness. Furthermore, the dual discriminators D1 and D2 also effectively prevent the negative migration phenomenon. The proposed DMCA-GAN obtained a DSC of 90.53% on the Medical Segmentation Decathlon (MSD) dataset with tenfold cross-validation, which is superior to the backbone by 3.78%.

Mesoscale quantification of grain boundary character and local plasticity in hydrogen-assisted intergranular and intergranular-like cracking paths in tempered martensitic steel

The effects of the misorientation of prior austenite grain boundary (PAGB) segments on the local plasticity evolution in intergranular (IG) and IG-like fractures were investigated using a tempered martensitic steel. When a diffusible hydrogen content of 6.3 mass ppm was introduced, IG fractures with smooth fracture surfaces occurred in the elastic regime of the tensile test. According to the grain orientation spread (GOS) distribution, local plasticity did not contribute to the mesoscopic (block size) scale, irrespective of the PAGB segment misorientation. With a relatively low hydrogen content (4.0 mass ppm), an IG-like feature appeared on the fracture surface, which included plasticity-related traces such as tear ridges or nanovoids. The GOS values near the IG-like cracks were higher than those near the IG cracks and exhibited significant misorientation dependence. Specifically, at the low-angle PAGB segments, the GOS values of the grains neighboring the IG-like cracks increased with a decreasing misorientation. On the other hand, most of the cracked high-angle PAGB segments exhibited nearly constant GOS values. Remarkably, a Σ3 PAGB segment exhibited an abnormally high GOS value. These facts and corresponding microstructural observation results indicate that the low angle and Σ3 PAGB segments allow crack-tip blunting before crack growth.