Precise Segmentation Research Articles

Multi scale multi attention network for blood vessel segmentation in fundus images

Precise segmentation of retinal vasculature is crucial for the early detection, diagnosis, and treatment of vision-threatening ailments. However, this task is challenging due to limited contextual information, variations in vessel thicknesses, the complexity of vessel structures, and the potential for confusion with lesions. In this paper, we introduce a novel approach, the MSMA Net model, which overcomes these challenges by replacing traditional convolution blocks and skip connections with an improved multi-scale squeeze and excitation block (MSSE Block) and Bottleneck residual paths (B-Res paths) with spatial attention blocks (SAB). Our experimental findings on publicly available datasets of fundus images, specifically DRIVE, STARE, CHASE_DB1, HRF and DR HAGIS consistently demonstrate that our approach outperforms other segmentation techniques, achieving higher accuracy, sensitivity, Dice score, and area under the receiver operator characteristic (AUC) in the segmentation of blood vessels with different thicknesses, even in situations involving diverse contextual information, the presence of coexisting lesions, and intricate vessel morphologies.

Research on Negative Road Obstacle Detection Based on Multimodal Feature Enhancement and Fusion

To address the issues of low recognition rates and poor detection accuracy for road negative obstacles caused by insufficient feature representation, we propose a novel detection framework: the Negative Road Obstacles Segmentation Network (NROSegNet). The detection performance of the algorithm is improved through a data enhancement strategy based on feature splicing and an adaptive feature enhancement module. Specifically, the data augmentation strategy introduces negative obstacle features into other datasets through geometric transformations and random splicing, effectively increasing the diversity of training data. This can solve the problem of an uneven distribution of data features while improving the performance of the model in capturing illumination changes and local details. The framework further adopts a dynamic multi-scale feature enhancement module to improve the perception of local details and global semantics. A robust multimodal data fusion mechanism and edge-aware optimization strategy are designed to effectively alleviate the problems of noise interference and boundary blur. The experimental results show that the NROSegNet proposed in this paper achieves 70.6 and 83.0 in mIoU and mF1, respectively, which is 2.8 percentage points and 2.9 percentage points higher than other methods. The results fully demonstrate its excellent performance in precise segmentation and boundary detail processing.

Colorectal cancer detection with enhanced precision using a hybrid supervised and unsupervised learning approach

The current work introduces the hybrid ensemble framework for the detection and segmentation of colorectal cancer. This framework will incorporate both supervised classification and unsupervised clustering methods to present more understandable and accurate diagnostic results. The method entails several steps with CNN models: ADa-22 and AD-22, transformer networks, and an SVM classifier, all inbuilt. The CVC ClinicDB dataset supports this process, containing 1650 colonoscopy images classified as polyps or non-polyps. The best performance in the ensembles was done by the AD-22 + Transformer + SVM model, with an AUC of 0.99, a training accuracy of 99.50%, and a testing accuracy of 99.00%. This group also saw a high accuracy of 97.50% for Polyps and 99.30% for Non-Polyps, together with a recall of 97.80% for Polyps and 98.90% for Non-Polyps, hence performing very well in identifying both cancerous and healthy regions. The framework proposed here uses K-means clustering in combination with the visualisation of bounding boxes, thereby improving segmentation and yielding a silhouette score of 0.73 with the best cluster configuration. It discusses how to combine feature interpretation challenges into medical imaging for accurate localization and precise segmentation of malignant regions. A good balance between performance and generalization shall be done by hyperparameter optimization-heavy learning rates; dropout rates and overfitting shall be suppressed effectively. The hybrid schema of this work treats the deficiencies of the previous approaches, such as incorporating CNN-based effective feature extraction, Transformer networks for developing attention mechanisms, and finally the fine decision boundary of the support vector machine. Further, we refine this process via unsupervised clustering for the purpose of enhancing the visualisation of such a procedure. Such a holistic framework, hence, further boosts classification and segmentation results by generating understandable outcomes for more rigorous benchmarking of detecting colorectal cancer and with higher reality towards clinical application feasibility.

Enhanced brain tumor detection and segmentation using densely connected convolutional networks with stacking ensemble learning.

- Brain tumors (BT), both benign and malignant, pose a substantial impact on human health and need precise and early detection for successful treatment. Analysing magnetic resonance imaging (MRI) image is a common method for BT diagnosis and segmentation, yet misdiagnoses yield effective medical responses, impacting patient survival rates. Recent technological advancements have popularized deep learning-based medical image analysis, leveraging transfer learning to reuse pre-trained models for various applications. BT segmentation with MRI remains challenging despite advancements in image acquisition techniques. Accurate detection and segmentation are essential for proper diagnosis and treatment planning. This study aims to enhance BT detection and segmentation accuracy and effectiveness of categorization through the implementation of an advanced stacking ensemble learning (SEL) approach. This study explores the efficiency of SEL architecture in augmenting the precision of BT segmentation. SEL, a prominent approach within the machine learning paradigm, combines the predictions of base-level models and improves the overall performance of predictions in order to reduce the errors and biases of each model. The proposed approach involves designing a stacked DenseNet201 as the meta-model called SEL-DenseNet201, complemented by six diverse base models such as mobile network version 3 (MobileNet-v3), 3-dimensional convolutional neural network (3D-CNN), visual geometry group network with 16 and 19 layers (VGG-16 and VGG-19), residual network with 50 layers (ResNet50), and Alex network (AlexNet). The strengths of the base models are calculated to capture distinct aspects of the BT MRI, aiming for enhanced segmentation performance. The proposed SEL-DenseNet201 is trained using BT MRI datasets. The augmentation techniques are applied to MRI scans to balance and enhance the model performance through the application of image enhancement and segmentation techniques. The proposed SEL-DenseNet201 achieves impressive results with an accuracy of 99.65% and a dice coefficient of 97.43%. These outcomes underscore the superiority of the proposed model over existing approaches. This study holds the potential to be an initial screening approach for early BT detection, with a high success rate.

Wound Segmentation with U-Net Using a Dual Attention Mechanism and Transfer Learning.

Accurate wound segmentation is crucial for the precise diagnosis and treatment of various skin conditions through image analysis. In this paper, we introduce a novel dual attention U-Net model designed for precise wound segmentation. Our proposed architecture integrates two widely used deep learning models, VGG16 and U-Net, incorporating dual attention mechanisms to focus on relevant regions within the wound area. Initially trained on diabetic foot ulcer images, we fine-tuned the model to acute and chronic wound images and conducted a comprehensive comparison with other state-of-the-art models. The results highlight the superior performance of our proposed dual attention model, achieving a Dice coefficient and IoU of 94.1% and 89.3%, respectively, on the test set. This underscores the robustness of our method and its capacity to generalize effectively to new data.

Standard observation plane annotation for diagnosis of ossicular chain in the temporal bone CT images.

Temporal bone CT is an essential technique for diagnosing ossicular chain trauma, and the location of standard observation planes (SOP) is the foundation of imaging diagnosis. The ossicular chain is small in volume, and there are about 11 standard observation planes for ossicular chain diagnosis, so it is a professional and time-consuming task to label SOPs accurately. An automatic annotation method of SOP is proposed. Firstly, an SOP-Graph model is introduced to represent the spatial relative position of different SOPs and the radiologists' SOP searching experience. Secondly, based on the precise segmentation of auditory ossicles, an SOP annotation algorithm was proposed to implement the parameter estimation of the SOP Graph. Finally, the evaluation of ossicular chain SOP localization was conducted, which converts the intensive SOP labeling task to automatic labeling and subjective verification. In the experiments, 610 CT images of temporal bone were automatically annotated, and the average success rate of annotation was 65.8% after being verified by a radiologist. By converting the manual annotation to automatic annotation and subjective verification, the annotation time of one case was reduced from 12 min to about 2 min. A data set that included 4414 SOP annotated slices was established.

Multiple token rearrangement Transformer network with explicit superpixel constraint for segmentation of echocardiography.

Diagnostic cardiologists have considerable clinical demand for precise segmentation of echocardiography to diagnose cardiovascular disease. The paradox is that manual segmentation of echocardiography is a time-consuming and operator-dependent task. Computer-aided segmentation can reduce the workflow greatly. However, it is challenging to segment multi-type echocardiography, which is reflected in differential anatomic structures, artifacts, and blurred borderline. This study proposes the multiple token rearrangement Transformer network (MTRT-Net) embedded in three novel modules to address the corresponding three challenges. First, the depthwise deformable attention module can extract flexible features to adapt to anatomic structures of echocardiography with different ages and diseases. Second, the superpixel supervised module can cluster similar features and keep discriminative features away to make the segmentation regions tend to be an entire body. The artifacts have the influence in separating the complete internal region. Third, the atrous affinity aggregation module can integrate affinity features near the borderline to judge the blurred regions. Overall, the three modules rearrange the relationships of tokens and broaden the diversity of features. Besides, the explicit constraint brought by the superpixel supervised module enhances the performance of fitting ability. This study has 13747 echocardiography to train and test the MTRT-Net. Abundant experiments also validate the performance of MTRT-Net. Therefore, MTRT-Net can assist the diagnostician in segmenting the echocardiography precisely.

MD-Unet for tobacco leaf disease spot segmentation based on multi-scale residual dilated convolutions

Identification and diagnosis of tobacco diseases are prerequisites for the scientific prevention and control of these ailments. To address the limitations of traditional methods, such as weak generalization and sensitivity to noise in segmenting tobacco leaf lesions, this study focused on four tobacco diseases: angular leaf spot, brown spot, wildfire disease, and frog eye disease. Building upon the Unet architecture, we developed the Multi-scale Residual Dilated Segmentation Model (MD-Unet) by enhancing the feature extraction module and integrating attention mechanisms. The results demonstrated that MD-Unet achieved 92.75%, 90.94%, 84.93%, and 91.81% for the lesion CPA, recall, IoU, and F1 metrics, respectively, with an overall Dice score of 94.67%. Furthermore, the model parameters, floating-point operations, and inference time per single image for MD-Unet were 4.65 × 107, 2.3392 × 1011, and 65.096 ms, respectively. Compared to Unet, PSP, DeepLab v3+, FCN, SegNet, UNET++, and DoubleU-Net, MD-Unet significantly improved accuracy while effectively managing model complexity, achieving optimal overall performance. This work provides the theoretical foundations and technical support for precise segmentation of tobacco lesions, with potential applications in the segmentation of other plant diseases.

Tissue optical clearing and expanding solution for improved nuclear segmentation

AbstractOver the recent years, fluorescence microscopy has seen substantial progress, transitioning from primarily qualitative analysis to a profoundly quantitative methodology, yet limitations in imaging depth and data quality persist due to tissue opacity. Tissue optical clearing (TOC) methodologies were implemented to overcome this hurdle, enabling the imaging and quantitative analysis of large tissue samples such as entire murine organs. Further improvements, including expansion microscopy merge TOC benefits with improved imaging resolution, converting cells and tissues into tissue‒gel hybrids. Yet, its intricate protocols lead to a remarkable, ∼10‒20× tissue volume expansion, making the acquisition of even thin sections exceedingly challenging in terms of time, labor, and data handling. Here, we present a novel solution, Improved Segmentation by gEntle Expansion (ISEE), for both TOC and gentle (1.6‒1.7×) tissue expansion for significantly improved precision of fluorescent signal segmentation, while maintaining tolerable increase during data acquisition in everyday laboratory practice. When applied to thick ∼40‒200 µm tissue slices, ISEE renders them transparent and ultimately expanded within minutes, in a simple, one‐step process compatible with immunohistochemistry as well as with all major murine organs.

ATEDU-NET: An Attention-Embedded Deep Unet for multi-disease diagnosis in chest X-ray images, breast ultrasound, and retina fundus.

In image segmentation for medical image analysis, effective upsampling is crucial for recovering spatial information lost during downsampling. This challenge becomes more pronounced when dealing with diverse medical image modalities, which can significantly impact model performance. Plain and standard skip connections, widely used in most models, often fall short of maintaining high segmentation accuracy across different modalities, because essential spatial information transferred from the encoder to the decoder is lost. Inspired by these limitations, the Attention-Embedded Deep UNet (ATEDU-Net) is presented here, an innovative framework designed and tested on diverse medical image modalities, including X-ray, breast ultrasound, and retinal fundus images. ATEDU-Net features a unique architecture that combines progressive context refinement modules (PCRM) and global context modules (GCM) within a U-shaped network structure with Convgroup blocks which facilitate the integration of convolutional operations. This allows ATEDU-Net to autonomously capture rich spatial details and crucial contextual information from input images. The GCM captures essential contextual information for informed decision-making, while the PCRM enhances feature attention, resulting in precise and robust segmentation. The experimentation and analysis demonstrate that ATEDU-Net promises to be a powerful tool for medical professionals, aiding in the early detection of chest-related diseases, accurate localization of breast tumors, and early identification of eye diseases. This, in turn, contributes significantly to the formulation of optimal therapeutic strategies, enhancing patient care and outcomes. The versatility of ATEDU-Net is further highlighted by its ability to analyze medical images across various modalities, making it well-suited for the complex task of medical image analysis in diverse clinical scenarios.

GCTransNet: 3D mitochondrial instance segmentation based on Global Context Vision Transformers.

Mitochondria are double membrane-bound organelles essential for generating energy in eukaryotic cells. Mitochondria can be readily visualized in 3D using Volume Electron Microscopy (vEM), and accurate image segmentation is vital for quantitative analysis of mitochondrial morphology and function. To address the challenge of segmenting small mitochondrial compartments in vEM images, we propose an automated mitochondrial segmentation method called GCTransNet. This method employs grayscale migration technology to preprocess images, effectively reducing intensity distribution differences across EM images. By utilizing 3D Global Context Vision Transformers (GC-ViT) combined with global context self-attention modules and local self-attention modules, GCTransNet precisely models long-range and short-range spatial interactions. The long-range interactions enable the model to capture the global structural relationships within the mitochondrial segmentation network, while the short-range interactions refine local details and boundaries. In our approach, the encoder of the 3D U-Net network, a classical multi-scale learning architecture that retains high-resolution features through skip connections and combines multi-scale features for precise segmentation, is replaced by a 3D GC-ViT. The GC-ViT leverages shifted window-based self-attention, capturing long-range dependencies and offering improved segmentation accuracy compared to traditional U-Net encoders. In the MitoEM mitochondrial segmentation challenge, GCTransNet achieved state-of-the-art results, demonstrating its superiority in automated mitochondrial segmentation. The code and its documentation are publicly available at https://github.com/GanLab123/GCTransNet.

Weakly Supervised Nuclei Segmentation with Point-Guided Attention and Self-Supervised Pseudo-Labeling.

Due to the labor-intensive manual annotations for nuclei segmentation, point-supervised segmentation based on nuclei coordinate supervision has gained recognition in recent years. Despite great progress, two challenges hinder the performance of weakly supervised nuclei segmentation methods: (1) The stable and effective segmentation of adjacent cell nuclei remains an unresolved challenge. (2) Existing approaches rely solely on initial pseudo-labels generated from point annotations for training, and inaccurate labels may lead the model to assimilate a considerable amount of noise information, thereby diminishing performance. To address these issues, we propose a method based on center-point prediction and pseudo-label updating for precise nuclei segmentation. First, we devise a Gaussian kernel mechanism that employs multi-scale Gaussian masks for multi-branch center-point prediction. The generated center points are utilized by the segmentation module to facilitate the effective separation of adjacent nuclei. Next, we introduce a point-guided attention mechanism that concentrates the segmentation module's attention around authentic point labels, reducing the noise impact caused by pseudo-labels. Finally, a label updating mechanism based on the exponential moving average (EMA) and k-means clustering is introduced to enhance the quality of pseudo-labels. The experimental results on three public datasets demonstrate that our approach has achieved state-of-the-art performance across multiple metrics. This method can significantly reduce annotation costs and reliance on clinical experts, facilitating large-scale dataset training and promoting the adoption of automated analysis in clinical applications.

Automatic segmentation of MRI images for brain radiotherapy planning using deep ensemble learning

Backgroundand Purpose:This study aimed to develop and evaluate an efficient method to automatically segment T1- and T2-weighted brain magnetic resonance imaging (MRI) images. We specifically compared the segmentation performance of individual convolutional neural network (CNN) models against an ensemble approach to advance the accuracy of MRI-guided radiotherapy (RT) planning.Materials and Methods. The evaluation was conducted on a private clinical dataset and a publicly available dataset (HaN-Seg). Anonymized MRI data from 55 brain cancer patients, including T1-weighted, T1-weighted with contrast, and T2-weighted images, were used in the clinical dataset. We employed an EDL strategy that integrated five independently trained 2D neural networks, each tailored for precise segmentation of tumors and organs at risk (OARs) in the MRI scans. Class probabilities were obtained by averaging the final layer activations (Softmax outputs) from the five networks using a weighted-average method, which were then converted into discrete labels. Segmentation performance was evaluated using the Dice similarity coefficient (DSC) and Hausdorff distance at 95% (HD95). The EDL model was also tested on the HaN-Seg public dataset for comparison.Results. The EDL model demonstrated superior segmentation performance on both the clinical and public datasets. For the clinical dataset, the ensemble approach achieved an average DSC of 0.7 ± 0.2 and HD95 of 4.5 ± 2.5 mm across all segmentations, significantly outperforming individual networks which yielded DSC values ≤0.6 and HD95 values ≥14 mm. Similar improvements were observed in the HaN-Seg public dataset.Conclusions. Our study shows that the EDL model consistently outperforms individual CNN networks in both clinical and public datasets, demonstrating the potential of ensemble learning to enhance segmentation accuracy. These findings underscore the value of the EDL approach for clinical applications, particularly in MRI-guided RT planning.

GLAC-Unet: Global-Local Active Contour Loss with an Efficient U-Shaped Architecture for Multiclass Medical Image Segmentation.

The field of medical image segmentation powered by deep learning has recently received substantial attention, with a significant focus on developing novel architectures and designing effective loss functions. Traditional loss functions, such as Dice loss and Cross-Entropy loss, predominantly rely on global metrics to compare predictions with labels. However, these global measures often struggle to address challenges such as occlusion and nonuni-form intensity. To overcome these issues, in this study, we propose a novel loss function, termed Global-Local Active Contour (GLAC) loss, which integrates both global and local image features, reformulated within the Mumford-Shah framework and extended for multiclass segmentation. This approach enables the neural network model to be trained end-to-end while simultaneously segmenting multiple classes. In addition to this, we enhance the U-Net architecture by incorporating Dense Layers, Convolutional Block Attention Modules, and DropBlock. These improvements enable the model to more effectively combine contextual information across layers, capture richer semantic details, and mitigate overfitting, resulting in more precise segmentation outcomes. We validate our proposed method, namely GLAC-Unet, which utilizes the GLAC loss in conjunction with our modified U-shaped architecture, on three biomedical segmentation datasets that span a range of modalities, including two-dimensional and three-dimensional images, such as dermoscopy, cardiac magnetic resonance imaging, and brain magnetic resonance imaging. Extensive experiments demonstrate the promising performance of our approach, achieving a Dice score (DSC) of 0.9125 on the ISIC-2018 dataset, 0.9260 on the Automated Cardiac Diagnosis Challenge (ACDC) 2017, and 0.927 on the Infant Brain MRI Segmentation Challenge 2019. Furthermore, statistical significance testing with p-values consistently smaller than 0.05 on the ISIC-2018 and ACDC datasets confirms the superior performance of the proposed method compared to other state-of-the-art models. These results highlight the robustness and effectiveness of our multiclass segmentation technique, underscoring its potential for biomedical image analysis. Our code will be made available at https://github.com/minhnhattrinh312/Active-Contour-Loss-based-on-Global-and-Local-Intensity.

Integrating attention mechanism and boundary detection for building segmentation from remote sensing images.

Accurate building segmentation has become critical in various fields such as urban management, urban planning, mapping, and navigation. With the increasing diversity in the number, size, and shape of buildings, convolutional neural networks have been used to segment and extract buildings from such images, resulting in increased efficiency and utilization of image features. We propose a building semantic segmentation method to improve the traditional Unet convolutional neural network by integrating attention mechanism and boundary detection. The attention mechanism module combines attention in the channel and spatial dimensions. The module captures image feature information in the channel dimension using a one-dimensional convolutional cross-channel method and automatically adjusts the cross-channel dimension using adaptive convolutional kernel size. Additionally, a weighted boundary loss function is designed to replace the traditional semantic segmentation cross-entropy loss to detect the boundary of a building. The loss function optimizes the extraction of building boundaries in backpropagation, ensuring the integrity of building boundary extraction in the shadow part. Experimental results show that the proposed model AMBDNet achieves high-performance metrics, including a recall rate of 0.9046, an IoU of 0.7797, and a pixel accuracy of 0.9140 on high-resolution remote sensing images, demonstrating its robustness and effectiveness in precise building segmentation. Experimental results further indicate that AMBDNet improves the single-class recall of buildings by 0.0322 and the single-class pixel accuracy by 0.0169 in the high-resolution remote sensing image recognition task.

MSEANet: Multi-Scale Selective Edge Aware Network for Polyp Segmentation

The colonoscopy procedure heavily relies on the operator’s expertise, underscoring the importance of automated polyp segmentation techniques in enhancing the efficiency and accuracy of colorectal cancer diagnosis. Nevertheless, achieving precise segmentation remains a significant challenge due to the high visual similarity between polyps and their backgrounds, blurred boundaries, and complex localization. To address these challenges, a Multi-scale Selective Edge-Aware Network has been proposed to facilitate polyp segmentation. The model consists of three key components: (1) an Edge Feature Extractor (EFE) that captures polyp edge features with precision during the initial encoding phase, (2) the Cross-layer Context Fusion (CCF) block designed to extract and integrate multi-scale contextual information from diverse receptive fields, and (3) the Selective Edge Aware (SEA) module that enhances sensitivity to high-frequency edge details during the decoding phase, thereby improving edge preservation and segmentation accuracy. The effectiveness of our model has been rigorously validated on the Kvasir-SEG, Kvasir-Sessile, and BKAI datasets, achieving mean Dice scores of 91.92%, 82.10%, and 92.24%, respectively, on the test sets.

Improving 3D deep learning segmentation with biophysically motivated cell synthesis

Biomedical research increasingly relies on three-dimensional (3D) cell culture models and artificial-intelligence-based analysis can potentially facilitate a detailed and accurate feature extraction on a single-cell level. However, this requires for a precise segmentation of 3D cell datasets, which in turn demands high-quality ground truth for training. Manual annotation, the gold standard for ground truth data, is too time-consuming and thus not feasible for the generation of large 3D training datasets. To address this, we present a framework for generating 3D training data, which integrates biophysical modeling for realistic cell shape and alignment. Our approach allows the in silico generation of coherent membrane and nuclei signals, that enable the training of segmentation models utilizing both channels for improved performance. Furthermore, we present a generative adversarial network (GAN) training scheme that generates not only image data but also matching labels. Quantitative evaluation shows superior performance of biophysical motivated synthetic training data, even outperforming manual annotation and pretrained models. This underscores the potential of incorporating biophysical modeling for enhancing synthetic training data quality.

Deep CNN-based semi-supervised learning approach for identifying and segmenting corrosion in hydraulic steel and water resources infrastructure

The United States faces significant challenges due to corrosion, with its impact on military and civilian infrastructure incurring over $20 billion in annual maintenance costs. The damage due to corrosion is profound, threatening structural safety, reducing esthetic value, and leading to costly repairs. To mitigate these effects, the Unified Facilities Criteria and Unified Facilities Guidance Specifications advise the use of protective coatings on metal surfaces. Early corrosion detection is crucial for maintaining structural integrity and minimizing maintenance costs. Recent breakthroughs in artificial intelligence and deep learning, including accurate corrosion classification, have significantly revolutionized the detection and management of corrosion. Despite these advancements, automatic corrosion segmentation in civil infrastructure remains challenging due to the scarcity of images and the labor-intensive annotation process. Moreover, existing segmentation methods are unable to manage the complexities that come with high-resolution corrosion images. This paper proposes a novel, semi-supervised, convolutional neural network-based image segmentation method for the automatic identification and segmentation of corrosion on coated steel surfaces, using both unlabeled and labeled corrosion images and leveraging the mean teacher model. The proposed novel method involves three steps: (1) utilizing high-resolution digital microscopy to capture detailed images and dividing them into manageable patches; (2) applying a semi-supervised learning approach, leveraging unlabeled corrosion images for enhanced segmentation precision; and (3) employing a smoothing module to improve the continuity of information. The proposed corrosion detection method has demonstrated promising performance with only 67% labeled data, achieving mean precision, recall, F-1 measure, and intersection over union of 90.0%, 96.2%, 92.7%, and 87.1%, respectively. Even with just 33% labeled data, the method maintains strong performance when compared to fully supervised deep learning models. This demonstrates a substantial data resource saving while ensuring accurate and reliable corrosion detection, which is crucial for infrastructure health monitoring. The successful validation of this approach provides a method that dramatically reduces the amount of visual data required to generate a reliable model.

CGV-Net: Tunnel Lining Crack Segmentation Method Based on Graph Convolution Guided Transformer

Lining cracking is among the most prevalent forms of tunnel distress, posing significant threats to tunnel operations and vehicular safety. The segmentation of tunnel lining cracks is often hindered by the influence of complex environmental factors, which makes relying solely on local feature extraction insufficient for achieving high segmentation accuracy. To address this issue, this study proposes CGV-Net (CNN, GNN, and ViT networks), a novel tunnel crack segmentation network model that integrates convolutional neural networks (CNNs), graph neural networks (GNNs), and Vision Transformers (ViTs). By fostering information exchange among local features, the model enhances comprehension of the global structural patterns of cracks and improves inference capabilities in recognizing intricate crack configurations. This approach effectively addresses the challenge of modeling contextual information in crack feature extraction. Additionally, the Detailed-Macro Feature Fusion (DMFF) module enables multi-scale feature integration by combining detailed and coarse-grained features, mitigating the significant feature loss encountered during the encoding and decoding stages, and further improving segmentation precision. To overcome the limitations of existing public datasets, which often feature a narrow range of crack types and simplistic backgrounds, this study introduces TunnelCrackDB, a dataset encompassing diverse crack types and complex backgrounds.Experimental evaluations on both the public Crack dataset and the newly developed TunnelCrackDB demonstrate the efficacy of CGV-Net. On the Crack dataset, CGV-Net achieves accuracy, recall, and F1 scores of 73.27% and 57.32%, respectively. On TunnelCrackDB, CGV-Net attains accuracy, recall, and F1 scores of 81.15%, 83.54%, and 82.33%, respectively, showcasing its superior performance in challenging segmentation tasks.

Deep learning-integrated MRI brain tumor analysis: feature extraction, segmentation, and Survival Prediction using Replicator and volumetric networks

The most prevalent form of malignant tumors that originate in the brain are known as gliomas. In order to diagnose, treat, and identify risk factors, it is crucial to have precise and resilient segmentation of the tumors, along with an estimation of the patients’ overall survival rate. Therefore, we have introduced a deep learning approach that employs a combination of MRI scans to accurately segment brain tumors and predict survival in patients with gliomas. To ensure strong and reliable tumor segmentation, we employ 2D volumetric convolution neural network architectures that utilize a majority rule. This method helps to significantly decrease model bias and improve performance. Additionally, in order to predict survival rates, we extract radiomic features from the tumor regions that have been segmented, and then use a Deep Learning Inspired 3D replicator neural network to identify the most effective features. The model presented in this study was successful in segmenting brain tumors and predicting the outcome of enhancing tumor and real enhancing tumor. The model was evaluated using the BRATS2020 benchmarks dataset, and the obtained results are quite satisfactory and promising.