Medical image segmentation is of critical importance in the domain of contemporary medical imaging.However,U-Net and its variants exhibit limitations in capturing complex nonlinear patterns and global contextual infor...Medical image segmentation is of critical importance in the domain of contemporary medical imaging.However,U-Net and its variants exhibit limitations in capturing complex nonlinear patterns and global contextual information.Although the subsequent U-KAN model enhances nonlinear representation capabilities,it still faces challenges such as gradient vanishing during deep network training and spatial detail loss during feature downsampling,resulting in insufficient segmentation accuracy for edge structures and minute lesions.To address these challenges,this paper proposes the RE-UKAN model,which innovatively improves upon U-KAN.Firstly,a residual network is introduced into the encoder to effectively mitigate gradient vanishing through cross-layer identity mappings,thus enhancing modelling capabilities for complex pathological structures.Secondly,Efficient Local Attention(ELA)is integrated to suppress spatial detail loss during downsampling,thereby improving the perception of edge structures and minute lesions.Experimental results on four public datasets demonstrate that RE-UKAN outperforms existing medical image segmentation methods across multiple evaluation metrics,with particularly outstanding performance on the TN-SCUI 2020 dataset,achieving IoU of 88.18%and Dice of 93.57%.Compared to the baseline model,it achieves improvements of 3.05%and 1.72%,respectively.These results fully demonstrate RE-UKAN’s superior detail retention capability and boundary recognition accuracy in complex medical image segmentation tasks,providing a reliable solution for clinical precision segmentation.展开更多
Recent advances in deep learning have significantly improved flood detection and segmentation from aerial and satellite imagery.However,conventional convolutional neural networks(CNNs)often struggle in complex flood s...Recent advances in deep learning have significantly improved flood detection and segmentation from aerial and satellite imagery.However,conventional convolutional neural networks(CNNs)often struggle in complex flood scenarios involving reflections,occlusions,or indistinct boundaries due to limited contextual modeling.To address these challenges,we propose a hybrid flood segmentation framework that integrates a Vision Transformer(ViT)encoder with a U-Net decoder,enhanced by a novel Flood-Aware Refinement Block(FARB).The FARB module improves boundary delineation and suppresses noise by combining residual smoothing with spatial-channel attention mechanisms.We evaluate our model on a UAV-acquired flood imagery dataset,demonstrating that the proposed ViTUNet+FARB architecture outperforms existing CNN and Transformer-based models in terms of accuracy and mean Intersection over Union(mIoU).Detailed ablation studies further validate the contribution of each component,confirming that the FARB design significantly enhances segmentation quality.To its better performance and computational efficiency,the proposed framework is well-suited for flood monitoring and disaster response applications,particularly in resource-constrained environments.展开更多
Background:Diabetic retinopathy(DR)is one of the primary causes of visual impairment globally,resulting from microvascular abnormalities in the retina.Accurate segmentation of retinal blood vessels from fundus images ...Background:Diabetic retinopathy(DR)is one of the primary causes of visual impairment globally,resulting from microvascular abnormalities in the retina.Accurate segmentation of retinal blood vessels from fundus images plays a pivotal role in the early diagnosis,progression monitoring,and treatment planning of DR and related ocular conditions.Traditional convolutional neural networks often struggle with capturing the intricate structures of thin vessels under varied illumination and contrast conditions.Methods:In this study,we propose an improved U-Net-based framework named MSAC U-Net,which enhances feature extraction and reconstruction through multiscale and attention-based modules.Specifically,the encoder replaces standard convolutions with a Multiscale Asymmetric Convolution(MSAC)block,incorporating parallel 1×n,n×1,and n×n kernels at different scales(3×3,5×5,7×7)to effectively capture fine-grained vascular structures.To further refine spatial representation,skip connections are utilized,and the decoder is augmented with dual activation strategies,Squeeze-and-Excitation blocks,and Convolutional Block Attention Modules for improved contextual understanding.Results:The model was evaluated on the publicly available DRIVE dataset.It achieved an accuracy of 96.48%,sensitivity of 88.31%,specificity of 97.90%,and an AUC of 98.59%,demonstrating superior performance compared to several state-of-the-art segmentation methods.Conclusion:The proposed MSAC U-Net provides a robust and accurate approach for retinal vessel segmentation,offering substantial clinical value in the early detection and management of diabetic retinopathy.Its design contributes to enhanced segmentation reliability and may serve as a foundation for broader applications in medical image analysis.展开更多
Near-infrared image sensors are widely used in fields such as material identification,machine vision,and autonomous driving.Lead sulfide colloidal quantum dot-based infrared photodiodes can be integrated with sil⁃icon...Near-infrared image sensors are widely used in fields such as material identification,machine vision,and autonomous driving.Lead sulfide colloidal quantum dot-based infrared photodiodes can be integrated with sil⁃icon-based readout circuits in a single step.Based on this,we propose a photodiode based on an n-i-p structure,which removes the buffer layer and further simplifies the manufacturing process of quantum dot image sensors,thus reducing manufacturing costs.Additionally,for the noise complexity in quantum dot image sensors when capturing images,traditional denoising and non-uniformity methods often do not achieve optimal denoising re⁃sults.For the noise and stripe-type non-uniformity commonly encountered in infrared quantum dot detector imag⁃es,a network architecture has been developed that incorporates multiple key modules.This network combines channel attention and spatial attention mechanisms,dynamically adjusting the importance of feature maps to en⁃hance the ability to distinguish between noise and details.Meanwhile,the residual dense feature fusion module further improves the network's ability to process complex image structures through hierarchical feature extraction and fusion.Furthermore,the pyramid pooling module effectively captures information at different scales,improv⁃ing the network's multi-scale feature representation ability.Through the collaborative effect of these modules,the network can better handle various mixed noise and image non-uniformity issues.Experimental results show that it outperforms the traditional U-Net network in denoising and image correction tasks.展开更多
Brain tumors require precise segmentation for diagnosis and treatment plans due to their complex morphology and heterogeneous characteristics.While MRI-based automatic brain tumor segmentation technology reduces the b...Brain tumors require precise segmentation for diagnosis and treatment plans due to their complex morphology and heterogeneous characteristics.While MRI-based automatic brain tumor segmentation technology reduces the burden on medical staff and provides quantitative information,existing methodologies and recent models still struggle to accurately capture and classify the fine boundaries and diverse morphologies of tumors.In order to address these challenges and maximize the performance of brain tumor segmentation,this research introduces a novel SwinUNETR-based model by integrating a new decoder block,the Hierarchical Channel-wise Attention Decoder(HCAD),into a powerful SwinUNETR encoder.The HCAD decoder block utilizes hierarchical features and channelspecific attention mechanisms to further fuse information at different scales transmitted from the encoder and preserve spatial details throughout the reconstruction phase.Rigorous evaluations on the recent BraTS GLI datasets demonstrate that the proposed SwinHCAD model achieved superior and improved segmentation accuracy on both the Dice score and HD95 metrics across all tumor subregions(WT,TC,and ET)compared to baseline models.In particular,the rationale and contribution of the model design were clarified through ablation studies to verify the effectiveness of the proposed HCAD decoder block.The results of this study are expected to greatly contribute to enhancing the efficiency of clinical diagnosis and treatment planning by increasing the precision of automated brain tumor segmentation.展开更多
Weakly supervised semantic segmentation(WSSS)is a tricky task,which only provides category information for segmentation prediction.Thus,the key stage of WSSS is to generate the pseudo labels.For convolutional neural n...Weakly supervised semantic segmentation(WSSS)is a tricky task,which only provides category information for segmentation prediction.Thus,the key stage of WSSS is to generate the pseudo labels.For convolutional neural network(CNN)based methods,in which class activation mapping(CAM)is proposed to obtain the pseudo labels,and only concentrates on the most discriminative parts.Recently,transformer-based methods utilize attention map from the multi-headed self-attention(MHSA)module to predict pseudo labels,which usually contain obvious background noise and incoherent object area.To solve the above problems,we use the Conformer as our backbone,which is a parallel network based on convolutional neural network(CNN)and Transformer.The two branches generate pseudo labels and refine them independently,and can effectively combine the advantages of CNN and Transformer.However,the parallel structure is not close enough in the information communication.Thus,parallel structure can result in poor details about pseudo labels,and the background noise still exists.To alleviate this problem,we propose enhancing convolution CAM(ECCAM)model,which have three improved modules based on enhancing convolution,including deeper stem(DStem),convolutional feed-forward network(CFFN)and feature coupling unit with convolution(FCUConv).The ECCAM could make Conformer have tighter interaction between CNN and Transformer branches.After experimental verification,the improved modules we propose can help the network perceive more local information from images,making the final segmentation results more refined.Compared with similar architecture,our modules greatly improve the semantic segmentation performance and achieve70.2%mean intersection over union(mIoU)on the PASCAL VOC 2012 dataset.展开更多
Automatic segmentation of landslides from remote sensing imagery is challenging because traditional machine learning and early CNN-based models often fail to generalize across heterogeneous landscapes,where segmentati...Automatic segmentation of landslides from remote sensing imagery is challenging because traditional machine learning and early CNN-based models often fail to generalize across heterogeneous landscapes,where segmentation maps contain sparse and fragmented landslide regions under diverse geographical conditions.To address these issues,we propose a lightweight dual-stream siamese deep learning framework that integrates optical and topographical data fusion with an adaptive decoder,guided multimodal fusion,and deep supervision.The framework is built upon the synergistic combination of cross-attention,gated fusion,and sub-pixel upsampling within a unified dual-stream architecture specifically optimized for landslide segmentation,enabling efficient context modeling and robust feature exchange between modalities.The decoder captures long-range context at deeper levels using lightweight cross-attention and refines spatial details at shallower levels through attention-gated skip fusion,enabling precise boundary delineation and fewer false positives.The gated fusion further enhances multimodal integration of optical and topographical cues,and the deep supervision stabilizes training and improves generalization.Moreover,to mitigate checkerboard artifacts,a learnable sub-pixel upsampling is devised to replace the traditional transposed convolution.Despite its compact design with fewer parameters,the model consistently outperforms state-of-the-art baselines.Experiments on two benchmark datasets,Landslide4Sense and Bijie,confirm the effectiveness of the framework.On the Bijie dataset,it achieves an F1-score of 0.9110 and an intersection over union(IoU)of 0.8839.These results highlight its potential for accurate large-scale landslide inventory mapping and real-time disaster response.The implementation is publicly available at https://github.com/mishaown/DiGATe-UNet-LandSlide-Segmentation(accessed on 3 November 2025).展开更多
The application of deep learning in fabric defect detection has become increasingly widespread.To address false positives and false negatives in fabric roll seam detection,and to improve automation efficiency and prod...The application of deep learning in fabric defect detection has become increasingly widespread.To address false positives and false negatives in fabric roll seam detection,and to improve automation efficiency and product quality,we propose the Multi-scale Context DeepLabV3+(MSC-DeepLabV3+),a semantic segmentation network designed for fabric roll seam detection,based on DeepLabV3+.The model improvements include enhancing the backbone performance through optimization of the UIB-MobileNetV2 network;designing the Dynamic Atrous and Sliding-window Fusion(DASF)module to improve adaptability to multi-scale seam structures with dynamic dilation rates and a sliding-window mechanism;and utilizing the Progressive Low-level Feature Fusion(PLFF)module to progressively restore seam boundary details via shallow feature fusion.Additionally,an enhanced 3-SE attention mechanism is employed,replacing the direct concatenation operation.Experimental results show thatMSCDeepLabV3+outperforms classical and recent segmentation models.Compared to DeepLabV3+with an Xception backbone,MSC-DeepLabV3+achieves a mean intersection over union(mIoU)of 92.30%and the boundary Fscore(BF)of 92.54%,representing improvements of 3.04%and 3.14%,respectively.Moreover,the model complexity is significantly reduced,with the model parameters(params)decreasing to 3.44M and Frames Per Second(FPS)increasing from 101 to 273,demonstrating its potential for deployment in resource-constrained industrial scenarios.展开更多
Recent studies indicate that millions of individuals suffer from renal diseases,with renal carcinoma,a type of kidney cancer,emerging as both a chronic illness and a significant cause of mortality.Magnetic Resonance I...Recent studies indicate that millions of individuals suffer from renal diseases,with renal carcinoma,a type of kidney cancer,emerging as both a chronic illness and a significant cause of mortality.Magnetic Resonance Imaging(MRI)and Computed Tomography(CT)have become essential tools for diagnosing and assessing kidney disorders.However,accurate analysis of thesemedical images is critical for detecting and evaluating tumor severity.This study introduces an integrated hybrid framework that combines three complementary deep learning models for kidney tumor segmentation from MRI images.The proposed framework fuses a customized U-Net and Mask R-CNN using a weighted scheme to achieve semantic and instance-level segmentation.The fused outputs are further refined through edge detection using Stochastic FeatureMapping Neural Networks(SFMNN),while volumetric consistency is ensured through Improved Mini-Batch K-Means(IMBKM)clustering integrated with an Encoder-Decoder Convolutional Neural Network(EDCNN).The outputs of these three stages are combined through a weighted fusion mechanism,with optimal weights determined empirically.Experiments on MRI scans from the TCGA-KIRC dataset demonstrate that the proposed hybrid framework significantly outperforms standalone models,achieving a Dice Score of 92.5%,an IoU of 87.8%,a Precision of 93.1%,a Recall of 90.8%,and a Hausdorff Distance of 2.8 mm.These findings validate that the weighted integration of complementary architectures effectively overcomes key limitations in kidney tumor segmentation,leading to improved diagnostic accuracy and robustness in medical image analysis.展开更多
Convolutional neural networks(CNNs)-based medical image segmentation technologies have been widely used in medical image segmentation because of their strong representation and generalization abilities.However,due to ...Convolutional neural networks(CNNs)-based medical image segmentation technologies have been widely used in medical image segmentation because of their strong representation and generalization abilities.However,due to the inability to effectively capture global information from images,CNNs can easily lead to loss of contours and textures in segmentation results.Notice that the transformer model can effectively capture the properties of long-range dependencies in the image,and furthermore,combining the CNN and the transformer can effectively extract local details and global contextual features of the image.Motivated by this,we propose a multi-branch and multi-scale attention network(M2ANet)for medical image segmentation,whose architecture consists of three components.Specifically,in the first component,we construct an adaptive multi-branch patch module for parallel extraction of image features to reduce information loss caused by downsampling.In the second component,we apply residual block to the well-known convolutional block attention module to enhance the network’s ability to recognize important features of images and alleviate the phenomenon of gradient vanishing.In the third component,we design a multi-scale feature fusion module,in which we adopt adaptive average pooling and position encoding to enhance contextual features,and then multi-head attention is introduced to further enrich feature representation.Finally,we validate the effectiveness and feasibility of the proposed M2ANet method through comparative experiments on four benchmark medical image segmentation datasets,particularly in the context of preserving contours and textures.展开更多
Convolutional neural network(CNN)with the encoder-decoder structure is popular in medical image segmentation due to its excellent local feature extraction ability but it faces limitations in capturing the global featu...Convolutional neural network(CNN)with the encoder-decoder structure is popular in medical image segmentation due to its excellent local feature extraction ability but it faces limitations in capturing the global feature.The transformer can extract the global information well but adapting it to small medical datasets is challenging and its computational complexity can be heavy.In this work,a serial and parallel network is proposed for the accurate 3D medical image segmentation by combining CNN and transformer and promoting feature interactions across various semantic levels.The core components of the proposed method include the cross window self-attention based transformer(CWST)and multi-scale local enhanced(MLE)modules.The CWST module enhances the global context understanding by partitioning 3D images into non-overlapping windows and calculating sparse global attention between windows.The MLE module selectively fuses features by computing the voxel attention between different branch features,and uses convolution to strengthen the dense local information.The experiments on the prostate,atrium,and pancreas MR/CT image datasets consistently demonstrate the advantage of the proposed method over six popular segmentation models in both qualitative evaluation and quantitative indexes such as dice similarity coefficient,Intersection over Union,95%Hausdorff distance and average symmetric surface distance.展开更多
Lower back pain is one of the most common medical problems in the world and it is experienced by a huge percentage of people everywhere.Due to its ability to produce a detailed view of the soft tissues,including the s...Lower back pain is one of the most common medical problems in the world and it is experienced by a huge percentage of people everywhere.Due to its ability to produce a detailed view of the soft tissues,including the spinal cord,nerves,intervertebral discs,and vertebrae,Magnetic Resonance Imaging is thought to be the most effective method for imaging the spine.The semantic segmentation of vertebrae plays a major role in the diagnostic process of lumbar diseases.It is difficult to semantically partition the vertebrae in Magnetic Resonance Images from the surrounding variety of tissues,including muscles,ligaments,and intervertebral discs.U-Net is a powerful deep-learning architecture to handle the challenges of medical image analysis tasks and achieves high segmentation accuracy.This work proposes a modified U-Net architecture namely MU-Net,consisting of the Meijering convolutional layer that incorporates the Meijering filter to perform the semantic segmentation of lumbar vertebrae L1 to L5 and sacral vertebra S1.Pseudo-colour mask images were generated and used as ground truth for training the model.The work has been carried out on 1312 images expanded from T1-weighted mid-sagittal MRI images of 515 patients in the Lumbar Spine MRI Dataset publicly available from Mendeley Data.The proposed MU-Net model for the semantic segmentation of the lumbar vertebrae gives better performance with 98.79%of pixel accuracy(PA),98.66%of dice similarity coefficient(DSC),97.36%of Jaccard coefficient,and 92.55%mean Intersection over Union(mean IoU)metrics using the mentioned dataset.展开更多
Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is cr...Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is crucial for computationally limited portable devices such as augmented reality and virtual reality.With the rapid advancements in deep learning,many network models have been developed specifically for eye image segmentation.Some methods divide the segmentation process into multiple stages to achieve model parameter miniaturization while enhancing output through post processing techniques to improve segmentation accuracy.These approaches significantly increase the inference time.Other networks adopt more complex encoding and decoding modules to achieve end-to-end output,which requires substantial computation.Therefore,balancing the model’s size,accuracy,and computational complexity is essential.To address these challenges,we propose a lightweight asymmetric UNet architecture and a projection loss function.We utilize ResNet-3 layer blocks to enhance feature extraction efficiency in the encoding stage.In the decoding stage,we employ regular convolutions and skip connections to upscale the feature maps from the latent space to the original image size,balancing the model size and segmentation accuracy.In addition,we leverage the geometric features of the eye region and design a projection loss function to further improve the segmentation accuracy without adding any additional inference computational cost.We validate our approach on the OpenEDS2019 dataset for virtual reality and achieve state-of-the-art performance with 95.33%mean intersection over union(mIoU).Our model has only 0.63M parameters and 350 FPS,which are 68%and 200%of the state-of-the-art model RITNet,respectively.展开更多
Medical image segmentation has become a cornerstone for many healthcare applications,allowing for the automated extraction of critical information from images such as Computed Tomography(CT)scans,Magnetic Resonance Im...Medical image segmentation has become a cornerstone for many healthcare applications,allowing for the automated extraction of critical information from images such as Computed Tomography(CT)scans,Magnetic Resonance Imaging(MRIs),and X-rays.The introduction of U-Net in 2015 has significantly advanced segmentation capabilities,especially for small datasets commonly found in medical imaging.Since then,various modifications to the original U-Net architecture have been proposed to enhance segmentation accuracy and tackle challenges like class imbalance,data scarcity,and multi-modal image processing.This paper provides a detailed review and comparison of several U-Net-based architectures,focusing on their effectiveness in medical image segmentation tasks.We evaluate performance metrics such as Dice Similarity Coefficient(DSC)and Intersection over Union(IoU)across different U-Net variants including HmsU-Net,CrossU-Net,mResU-Net,and others.Our results indicate that architectural enhancements such as transformers,attention mechanisms,and residual connections improve segmentation performance across diverse medical imaging applications,including tumor detection,organ segmentation,and lesion identification.The study also identifies current challenges in the field,including data variability,limited dataset sizes,and issues with class imbalance.Based on these findings,the paper suggests potential future directions for improving the robustness and clinical applicability of U-Net-based models in medical image segmentation.展开更多
Deep learning(DL),derived from the domain of Artificial Neural Networks(ANN),forms one of the most essential components of modern deep learning algorithms.DL segmentation models rely on layer-by-layer convolution-base...Deep learning(DL),derived from the domain of Artificial Neural Networks(ANN),forms one of the most essential components of modern deep learning algorithms.DL segmentation models rely on layer-by-layer convolution-based feature representation,guided by forward and backward propagation.Acritical aspect of this process is the selection of an appropriate activation function(AF)to ensure robustmodel learning.However,existing activation functions often fail to effectively address the vanishing gradient problem or are complicated by the need for manual parameter tuning.Most current research on activation function design focuses on classification tasks using natural image datasets such asMNIST,CIFAR-10,and CIFAR-100.To address this gap,this study proposesMed-ReLU,a novel activation function specifically designed for medical image segmentation.Med-ReLU prevents deep learning models fromsuffering dead neurons or vanishing gradient issues.It is a hybrid activation function that combines the properties of ReLU and Softsign.For positive inputs,Med-ReLU adopts the linear behavior of ReLU to avoid vanishing gradients,while for negative inputs,it exhibits the Softsign’s polynomial convergence,ensuring robust training and avoiding inactive neurons across the training set.The training performance and segmentation accuracy ofMed-ReLU have been thoroughly evaluated,demonstrating stable learning behavior and resistance to overfitting.It consistently outperforms state-of-the-art activation functions inmedical image segmentation tasks.Designed as a parameter-free function,Med-ReLU is simple to implement in complex deep learning architectures,and its effectiveness spans various neural network models and anomaly detection scenarios.展开更多
Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a chall...Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a challenge for accurate segmentation.In this paper,we propose a 3D semantic segmentation network for neuronal soma segmentation to address this issue.Using an encoding-decoding structure,we introduce a Multi-Scale feature extraction and Adaptive Weighting fusion module(MSAW)after each encoding block.The MSAW module can not only emphasize the fine structures via an upsampling strategy,but also provide pixel-wise weights to measure the importance of the multi-scale features.Additionally,a dynamic convolution instead of normal convolution is employed to better adapt the network to input data with different distributions.The proposed MSAW-based semantic segmentation network(MSAW-Net)was evaluated on three neuronal soma images from mouse brain and one neuronal soma image from macaque brain,demonstrating the efficiency of the proposed method.It achieved an F1 score of 91.8%on Fezf2-2A-CreER dataset,97.1%on LSL-H2B-GFP dataset,82.8%on Thy1-EGFP-Mline dataset,and 86.9%on macaque dataset,achieving improvements over the 3D U-Net model by 3.1%,3.3%,3.9%,and 2.3%,respectively.展开更多
Thyroid nodules,a common disorder in the endocrine system,require accurate segmentation in ultrasound images for effective diagnosis and treatment.However,achieving precise segmentation remains a challenge due to vari...Thyroid nodules,a common disorder in the endocrine system,require accurate segmentation in ultrasound images for effective diagnosis and treatment.However,achieving precise segmentation remains a challenge due to various factors,including scattering noise,low contrast,and limited resolution in ultrasound images.Although existing segmentation models have made progress,they still suffer from several limitations,such as high error rates,low generalizability,overfitting,limited feature learning capability,etc.To address these challenges,this paper proposes a Multi-level Relation Transformer-based U-Net(MLRT-UNet)to improve thyroid nodule segmentation.The MLRTUNet leverages a novel Relation Transformer,which processes images at multiple scales,overcoming the limitations of traditional encoding methods.This transformer integrates both local and global features effectively through selfattention and cross-attention units,capturing intricate relationships within the data.The approach also introduces a Co-operative Transformer Fusion(CTF)module to combine multi-scale features from different encoding layers,enhancing the model’s ability to capture complex patterns in the data.Furthermore,the Relation Transformer block enhances long-distance dependencies during the decoding process,improving segmentation accuracy.Experimental results showthat the MLRT-UNet achieves high segmentation accuracy,reaching 98.2% on the Digital Database Thyroid Image(DDT)dataset,97.8% on the Thyroid Nodule 3493(TG3K)dataset,and 98.2% on the Thyroid Nodule3K(TN3K)dataset.These findings demonstrate that the proposed method significantly enhances the accuracy of thyroid nodule segmentation,addressing the limitations of existing models.展开更多
Brain tumor segmentation from Magnetic Resonance Imaging(MRI)supports neurologists and radiologists in analyzing tumors and developing personalized treatment plans,making it a crucial yet challenging task.Supervised m...Brain tumor segmentation from Magnetic Resonance Imaging(MRI)supports neurologists and radiologists in analyzing tumors and developing personalized treatment plans,making it a crucial yet challenging task.Supervised models such as 3D U-Net perform well in this domain,but their accuracy significantly improves with appropriate preprocessing.This paper demonstrates the effectiveness of preprocessing in brain tumor segmentation by applying a pre-segmentation step based on the Generalized Gaussian Mixture Model(GGMM)to T1 contrastenhanced MRI scans from the BraTS 2020 dataset.The Expectation-Maximization(EM)algorithm is employed to estimate parameters for four tissue classes,generating a new pre-segmented channel that enhances the training and performance of the 3DU-Net model.The proposed GGMM+3D U-Net framework achieved a Dice coefficient of 0.88 for whole tumor segmentation,outperforming both the standard multiscale 3D U-Net(0.84)and MMU-Net(0.85).It also delivered higher Intersection over Union(IoU)scores compared to models trained without preprocessing or with simpler GMM-based segmentation.These results,supported by qualitative visualizations,suggest that GGMM-based preprocessing should be integrated into brain tumor segmentation pipelines to optimize performance.展开更多
This study presents an advanced method for post-mortem person identification using the segmentation of skeletal structures from chest X-ray images.The proposed approach employs the Attention U-Net architecture,enhance...This study presents an advanced method for post-mortem person identification using the segmentation of skeletal structures from chest X-ray images.The proposed approach employs the Attention U-Net architecture,enhanced with gated attention mechanisms,to refine segmentation by emphasizing spatially relevant anatomical features while suppressing irrelevant details.By isolating skeletal structures which remain stable over time compared to soft tissues,this method leverages bones as reliable biometric markers for identity verification.The model integrates custom-designed encoder and decoder blocks with attention gates,achieving high segmentation precision.To evaluate the impact of architectural choices,we conducted an ablation study comparing Attention U-Net with and without attentionmechanisms,alongside an analysis of data augmentation effects.Training and evaluation were performed on a curated chest X-ray dataset,with segmentation performance measured using Dice score,precision,and loss functions,achieving over 98% precision and 94% Dice score.The extracted bone structures were further processed to derive unique biometric patterns,enabling robust and privacy-preserving person identification.Our findings highlight the effectiveness of attentionmechanisms in improving segmentation accuracy and underscore the potential of chest bonebased biometrics in forensic and medical imaging.This work paves the way for integrating artificial intelligence into real-world forensic workflows,offering a non-invasive and reliable solution for post-mortem identification.展开更多
With the continuous development of artificial intelligence and machine learning techniques,there have been effective methods supporting the work of dermatologist in the field of skin cancer detection.However,object si...With the continuous development of artificial intelligence and machine learning techniques,there have been effective methods supporting the work of dermatologist in the field of skin cancer detection.However,object significant challenges have been presented in accurately segmenting melanomas in dermoscopic images due to the objects that could interfere human observations,such as bubbles and scales.To address these challenges,we propose a dual U-Net network framework for skin melanoma segmentation.In our proposed architecture,we introduce several innovative components that aim to enhance the performance and capabilities of the traditional U-Net.First,we establish a novel framework that links two simplified U-Nets,enabling more comprehensive information exchange and feature integration throughout the network.Second,after cascading the second U-Net,we introduce a skip connection between the decoder and encoder networks,and incorporate a modified receptive field block(MRFB),which is designed to capture multi-scale spatial information.Third,to further enhance the feature representation capabilities,we add a multi-path convolution block attention module(MCBAM)to the first two layers of the first U-Net encoding,and integrate a new squeeze-and-excitation(SE)mechanism with residual connections in the second U-Net.To illustrate the performance of our proposed model,we conducted comprehensive experiments on widely recognized skin datasets.On the ISIC-2017 dataset,the IoU value of our proposed model increased from 0.6406 to 0.6819 and the Dice coefficient increased from 0.7625 to 0.8023.On the ISIC-2018 dataset,the IoU value of proposed model also improved from 0.7138 to 0.7709,while the Dice coefficient increased from 0.8285 to 0.8665.Furthermore,the generalization experiments conducted on the jaw cyst dataset from Quzhou People’s Hospital further verified the outstanding segmentation performance of the proposed model.These findings collectively affirm the potential of our approach as a valuable tool in supporting clinical decision-making in the field of skin cancer detection,as well as advancing research in medical image analysis.展开更多
文摘Medical image segmentation is of critical importance in the domain of contemporary medical imaging.However,U-Net and its variants exhibit limitations in capturing complex nonlinear patterns and global contextual information.Although the subsequent U-KAN model enhances nonlinear representation capabilities,it still faces challenges such as gradient vanishing during deep network training and spatial detail loss during feature downsampling,resulting in insufficient segmentation accuracy for edge structures and minute lesions.To address these challenges,this paper proposes the RE-UKAN model,which innovatively improves upon U-KAN.Firstly,a residual network is introduced into the encoder to effectively mitigate gradient vanishing through cross-layer identity mappings,thus enhancing modelling capabilities for complex pathological structures.Secondly,Efficient Local Attention(ELA)is integrated to suppress spatial detail loss during downsampling,thereby improving the perception of edge structures and minute lesions.Experimental results on four public datasets demonstrate that RE-UKAN outperforms existing medical image segmentation methods across multiple evaluation metrics,with particularly outstanding performance on the TN-SCUI 2020 dataset,achieving IoU of 88.18%and Dice of 93.57%.Compared to the baseline model,it achieves improvements of 3.05%and 1.72%,respectively.These results fully demonstrate RE-UKAN’s superior detail retention capability and boundary recognition accuracy in complex medical image segmentation tasks,providing a reliable solution for clinical precision segmentation.
基金supported by the National Research Foundation of Korea(NRF)grant funded by theKorea government(MSIT)(No.RS-2024-00405278)partially supported by the Jeju Industry-University Convergence District Project for Promoting Industry-Campus Cooperationfunded by the Ministry of Trade,Industry and Energy(MOTIE,Korea)[Project Name:Jeju Industry-University Convergence District Project for Promoting Industry-Campus Cooperation/Project Number:P0029950].
文摘Recent advances in deep learning have significantly improved flood detection and segmentation from aerial and satellite imagery.However,conventional convolutional neural networks(CNNs)often struggle in complex flood scenarios involving reflections,occlusions,or indistinct boundaries due to limited contextual modeling.To address these challenges,we propose a hybrid flood segmentation framework that integrates a Vision Transformer(ViT)encoder with a U-Net decoder,enhanced by a novel Flood-Aware Refinement Block(FARB).The FARB module improves boundary delineation and suppresses noise by combining residual smoothing with spatial-channel attention mechanisms.We evaluate our model on a UAV-acquired flood imagery dataset,demonstrating that the proposed ViTUNet+FARB architecture outperforms existing CNN and Transformer-based models in terms of accuracy and mean Intersection over Union(mIoU).Detailed ablation studies further validate the contribution of each component,confirming that the FARB design significantly enhances segmentation quality.To its better performance and computational efficiency,the proposed framework is well-suited for flood monitoring and disaster response applications,particularly in resource-constrained environments.
基金supported by the Guangdong Basic and Applied Basic Research Foundation(2024A1515010987)the Medical Scientific Research Foundation of Guangdong Province(B2024035).
文摘Background:Diabetic retinopathy(DR)is one of the primary causes of visual impairment globally,resulting from microvascular abnormalities in the retina.Accurate segmentation of retinal blood vessels from fundus images plays a pivotal role in the early diagnosis,progression monitoring,and treatment planning of DR and related ocular conditions.Traditional convolutional neural networks often struggle with capturing the intricate structures of thin vessels under varied illumination and contrast conditions.Methods:In this study,we propose an improved U-Net-based framework named MSAC U-Net,which enhances feature extraction and reconstruction through multiscale and attention-based modules.Specifically,the encoder replaces standard convolutions with a Multiscale Asymmetric Convolution(MSAC)block,incorporating parallel 1×n,n×1,and n×n kernels at different scales(3×3,5×5,7×7)to effectively capture fine-grained vascular structures.To further refine spatial representation,skip connections are utilized,and the decoder is augmented with dual activation strategies,Squeeze-and-Excitation blocks,and Convolutional Block Attention Modules for improved contextual understanding.Results:The model was evaluated on the publicly available DRIVE dataset.It achieved an accuracy of 96.48%,sensitivity of 88.31%,specificity of 97.90%,and an AUC of 98.59%,demonstrating superior performance compared to several state-of-the-art segmentation methods.Conclusion:The proposed MSAC U-Net provides a robust and accurate approach for retinal vessel segmentation,offering substantial clinical value in the early detection and management of diabetic retinopathy.Its design contributes to enhanced segmentation reliability and may serve as a foundation for broader applications in medical image analysis.
基金Supported by the National key research and development program in the 14th five year plan 2021YFA1200700)the National Natural Science Foundation of China(62535018,62431025,62561160113)the Natural Science Foundation of Shanghai(23ZR1473400).
文摘Near-infrared image sensors are widely used in fields such as material identification,machine vision,and autonomous driving.Lead sulfide colloidal quantum dot-based infrared photodiodes can be integrated with sil⁃icon-based readout circuits in a single step.Based on this,we propose a photodiode based on an n-i-p structure,which removes the buffer layer and further simplifies the manufacturing process of quantum dot image sensors,thus reducing manufacturing costs.Additionally,for the noise complexity in quantum dot image sensors when capturing images,traditional denoising and non-uniformity methods often do not achieve optimal denoising re⁃sults.For the noise and stripe-type non-uniformity commonly encountered in infrared quantum dot detector imag⁃es,a network architecture has been developed that incorporates multiple key modules.This network combines channel attention and spatial attention mechanisms,dynamically adjusting the importance of feature maps to en⁃hance the ability to distinguish between noise and details.Meanwhile,the residual dense feature fusion module further improves the network's ability to process complex image structures through hierarchical feature extraction and fusion.Furthermore,the pyramid pooling module effectively captures information at different scales,improv⁃ing the network's multi-scale feature representation ability.Through the collaborative effect of these modules,the network can better handle various mixed noise and image non-uniformity issues.Experimental results show that it outperforms the traditional U-Net network in denoising and image correction tasks.
基金supported by Institute of Information&Communications Technology Planning&Evaluation(IITP)under the Metaverse Support Program to Nurture the Best Talents(IITP-2024-RS-2023-00254529)grant funded by the Korea government(MSIT).
文摘Brain tumors require precise segmentation for diagnosis and treatment plans due to their complex morphology and heterogeneous characteristics.While MRI-based automatic brain tumor segmentation technology reduces the burden on medical staff and provides quantitative information,existing methodologies and recent models still struggle to accurately capture and classify the fine boundaries and diverse morphologies of tumors.In order to address these challenges and maximize the performance of brain tumor segmentation,this research introduces a novel SwinUNETR-based model by integrating a new decoder block,the Hierarchical Channel-wise Attention Decoder(HCAD),into a powerful SwinUNETR encoder.The HCAD decoder block utilizes hierarchical features and channelspecific attention mechanisms to further fuse information at different scales transmitted from the encoder and preserve spatial details throughout the reconstruction phase.Rigorous evaluations on the recent BraTS GLI datasets demonstrate that the proposed SwinHCAD model achieved superior and improved segmentation accuracy on both the Dice score and HD95 metrics across all tumor subregions(WT,TC,and ET)compared to baseline models.In particular,the rationale and contribution of the model design were clarified through ablation studies to verify the effectiveness of the proposed HCAD decoder block.The results of this study are expected to greatly contribute to enhancing the efficiency of clinical diagnosis and treatment planning by increasing the precision of automated brain tumor segmentation.
文摘Weakly supervised semantic segmentation(WSSS)is a tricky task,which only provides category information for segmentation prediction.Thus,the key stage of WSSS is to generate the pseudo labels.For convolutional neural network(CNN)based methods,in which class activation mapping(CAM)is proposed to obtain the pseudo labels,and only concentrates on the most discriminative parts.Recently,transformer-based methods utilize attention map from the multi-headed self-attention(MHSA)module to predict pseudo labels,which usually contain obvious background noise and incoherent object area.To solve the above problems,we use the Conformer as our backbone,which is a parallel network based on convolutional neural network(CNN)and Transformer.The two branches generate pseudo labels and refine them independently,and can effectively combine the advantages of CNN and Transformer.However,the parallel structure is not close enough in the information communication.Thus,parallel structure can result in poor details about pseudo labels,and the background noise still exists.To alleviate this problem,we propose enhancing convolution CAM(ECCAM)model,which have three improved modules based on enhancing convolution,including deeper stem(DStem),convolutional feed-forward network(CFFN)and feature coupling unit with convolution(FCUConv).The ECCAM could make Conformer have tighter interaction between CNN and Transformer branches.After experimental verification,the improved modules we propose can help the network perceive more local information from images,making the final segmentation results more refined.Compared with similar architecture,our modules greatly improve the semantic segmentation performance and achieve70.2%mean intersection over union(mIoU)on the PASCAL VOC 2012 dataset.
基金funded by the National Natural Science Foundation of China,grant number 62262045the Fundamental Research Funds for the Central Universities,grant number 2023CDJYGRH-YB11the Open Funding of SUGON Industrial Control and Security Center,grant number CUIT-SICSC-2025-03.
文摘Automatic segmentation of landslides from remote sensing imagery is challenging because traditional machine learning and early CNN-based models often fail to generalize across heterogeneous landscapes,where segmentation maps contain sparse and fragmented landslide regions under diverse geographical conditions.To address these issues,we propose a lightweight dual-stream siamese deep learning framework that integrates optical and topographical data fusion with an adaptive decoder,guided multimodal fusion,and deep supervision.The framework is built upon the synergistic combination of cross-attention,gated fusion,and sub-pixel upsampling within a unified dual-stream architecture specifically optimized for landslide segmentation,enabling efficient context modeling and robust feature exchange between modalities.The decoder captures long-range context at deeper levels using lightweight cross-attention and refines spatial details at shallower levels through attention-gated skip fusion,enabling precise boundary delineation and fewer false positives.The gated fusion further enhances multimodal integration of optical and topographical cues,and the deep supervision stabilizes training and improves generalization.Moreover,to mitigate checkerboard artifacts,a learnable sub-pixel upsampling is devised to replace the traditional transposed convolution.Despite its compact design with fewer parameters,the model consistently outperforms state-of-the-art baselines.Experiments on two benchmark datasets,Landslide4Sense and Bijie,confirm the effectiveness of the framework.On the Bijie dataset,it achieves an F1-score of 0.9110 and an intersection over union(IoU)of 0.8839.These results highlight its potential for accurate large-scale landslide inventory mapping and real-time disaster response.The implementation is publicly available at https://github.com/mishaown/DiGATe-UNet-LandSlide-Segmentation(accessed on 3 November 2025).
文摘The application of deep learning in fabric defect detection has become increasingly widespread.To address false positives and false negatives in fabric roll seam detection,and to improve automation efficiency and product quality,we propose the Multi-scale Context DeepLabV3+(MSC-DeepLabV3+),a semantic segmentation network designed for fabric roll seam detection,based on DeepLabV3+.The model improvements include enhancing the backbone performance through optimization of the UIB-MobileNetV2 network;designing the Dynamic Atrous and Sliding-window Fusion(DASF)module to improve adaptability to multi-scale seam structures with dynamic dilation rates and a sliding-window mechanism;and utilizing the Progressive Low-level Feature Fusion(PLFF)module to progressively restore seam boundary details via shallow feature fusion.Additionally,an enhanced 3-SE attention mechanism is employed,replacing the direct concatenation operation.Experimental results show thatMSCDeepLabV3+outperforms classical and recent segmentation models.Compared to DeepLabV3+with an Xception backbone,MSC-DeepLabV3+achieves a mean intersection over union(mIoU)of 92.30%and the boundary Fscore(BF)of 92.54%,representing improvements of 3.04%and 3.14%,respectively.Moreover,the model complexity is significantly reduced,with the model parameters(params)decreasing to 3.44M and Frames Per Second(FPS)increasing from 101 to 273,demonstrating its potential for deployment in resource-constrained industrial scenarios.
基金funded by the Ongoing Research Funding Program-Research Chairs(ORF-RC-2025-2400),King Saud University,Riyadh,Saudi Arabia。
文摘Recent studies indicate that millions of individuals suffer from renal diseases,with renal carcinoma,a type of kidney cancer,emerging as both a chronic illness and a significant cause of mortality.Magnetic Resonance Imaging(MRI)and Computed Tomography(CT)have become essential tools for diagnosing and assessing kidney disorders.However,accurate analysis of thesemedical images is critical for detecting and evaluating tumor severity.This study introduces an integrated hybrid framework that combines three complementary deep learning models for kidney tumor segmentation from MRI images.The proposed framework fuses a customized U-Net and Mask R-CNN using a weighted scheme to achieve semantic and instance-level segmentation.The fused outputs are further refined through edge detection using Stochastic FeatureMapping Neural Networks(SFMNN),while volumetric consistency is ensured through Improved Mini-Batch K-Means(IMBKM)clustering integrated with an Encoder-Decoder Convolutional Neural Network(EDCNN).The outputs of these three stages are combined through a weighted fusion mechanism,with optimal weights determined empirically.Experiments on MRI scans from the TCGA-KIRC dataset demonstrate that the proposed hybrid framework significantly outperforms standalone models,achieving a Dice Score of 92.5%,an IoU of 87.8%,a Precision of 93.1%,a Recall of 90.8%,and a Hausdorff Distance of 2.8 mm.These findings validate that the weighted integration of complementary architectures effectively overcomes key limitations in kidney tumor segmentation,leading to improved diagnostic accuracy and robustness in medical image analysis.
基金supported by the Natural Science Foundation of the Anhui Higher Education Institutions of China(Grant Nos.2023AH040149 and 2024AH051915)the Anhui Provincial Natural Science Foundation(Grant No.2208085MF168)+1 种基金the Science and Technology Innovation Tackle Plan Project of Maanshan(Grant No.2024RGZN001)the Scientific Research Fund Project of Anhui Medical University(Grant No.2023xkj122).
文摘Convolutional neural networks(CNNs)-based medical image segmentation technologies have been widely used in medical image segmentation because of their strong representation and generalization abilities.However,due to the inability to effectively capture global information from images,CNNs can easily lead to loss of contours and textures in segmentation results.Notice that the transformer model can effectively capture the properties of long-range dependencies in the image,and furthermore,combining the CNN and the transformer can effectively extract local details and global contextual features of the image.Motivated by this,we propose a multi-branch and multi-scale attention network(M2ANet)for medical image segmentation,whose architecture consists of three components.Specifically,in the first component,we construct an adaptive multi-branch patch module for parallel extraction of image features to reduce information loss caused by downsampling.In the second component,we apply residual block to the well-known convolutional block attention module to enhance the network’s ability to recognize important features of images and alleviate the phenomenon of gradient vanishing.In the third component,we design a multi-scale feature fusion module,in which we adopt adaptive average pooling and position encoding to enhance contextual features,and then multi-head attention is introduced to further enrich feature representation.Finally,we validate the effectiveness and feasibility of the proposed M2ANet method through comparative experiments on four benchmark medical image segmentation datasets,particularly in the context of preserving contours and textures.
基金National Key Research and Development Program of China,Grant/Award Number:2018YFE0206900China Postdoctoral Science Foundation,Grant/Award Number:2023M731204+2 种基金The Open Project of Key Laboratory for Quality Evaluation of Ultrasound Surgical Equipment of National Medical Products Administration,Grant/Award Number:SMDTKL-2023-1-01The Hubei Province Key Research and Development Project,Grant/Award Number:2023BCB007CAAI-Huawei MindSpore Open Fund。
文摘Convolutional neural network(CNN)with the encoder-decoder structure is popular in medical image segmentation due to its excellent local feature extraction ability but it faces limitations in capturing the global feature.The transformer can extract the global information well but adapting it to small medical datasets is challenging and its computational complexity can be heavy.In this work,a serial and parallel network is proposed for the accurate 3D medical image segmentation by combining CNN and transformer and promoting feature interactions across various semantic levels.The core components of the proposed method include the cross window self-attention based transformer(CWST)and multi-scale local enhanced(MLE)modules.The CWST module enhances the global context understanding by partitioning 3D images into non-overlapping windows and calculating sparse global attention between windows.The MLE module selectively fuses features by computing the voxel attention between different branch features,and uses convolution to strengthen the dense local information.The experiments on the prostate,atrium,and pancreas MR/CT image datasets consistently demonstrate the advantage of the proposed method over six popular segmentation models in both qualitative evaluation and quantitative indexes such as dice similarity coefficient,Intersection over Union,95%Hausdorff distance and average symmetric surface distance.
文摘Lower back pain is one of the most common medical problems in the world and it is experienced by a huge percentage of people everywhere.Due to its ability to produce a detailed view of the soft tissues,including the spinal cord,nerves,intervertebral discs,and vertebrae,Magnetic Resonance Imaging is thought to be the most effective method for imaging the spine.The semantic segmentation of vertebrae plays a major role in the diagnostic process of lumbar diseases.It is difficult to semantically partition the vertebrae in Magnetic Resonance Images from the surrounding variety of tissues,including muscles,ligaments,and intervertebral discs.U-Net is a powerful deep-learning architecture to handle the challenges of medical image analysis tasks and achieves high segmentation accuracy.This work proposes a modified U-Net architecture namely MU-Net,consisting of the Meijering convolutional layer that incorporates the Meijering filter to perform the semantic segmentation of lumbar vertebrae L1 to L5 and sacral vertebra S1.Pseudo-colour mask images were generated and used as ground truth for training the model.The work has been carried out on 1312 images expanded from T1-weighted mid-sagittal MRI images of 515 patients in the Lumbar Spine MRI Dataset publicly available from Mendeley Data.The proposed MU-Net model for the semantic segmentation of the lumbar vertebrae gives better performance with 98.79%of pixel accuracy(PA),98.66%of dice similarity coefficient(DSC),97.36%of Jaccard coefficient,and 92.55%mean Intersection over Union(mean IoU)metrics using the mentioned dataset.
基金supported by the HFIPS Director’s Foundation(YZJJ202207-TS),the National Natural Science Foundation of China(82371931)the Natural Science Foundation of Anhui Province(2008085MC69)+3 种基金the Natural Science Foundation of Hefei City(2021033)the General Scientific Research Project of Anhui Provincial Health Commission(AHWJ2021b150)the Collaborative Innovation Program of Hefei Science Center,CAS(2021HSC-CIP013)the Anhui Province Key Research and Development Project(202204295107020004).
文摘Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is crucial for computationally limited portable devices such as augmented reality and virtual reality.With the rapid advancements in deep learning,many network models have been developed specifically for eye image segmentation.Some methods divide the segmentation process into multiple stages to achieve model parameter miniaturization while enhancing output through post processing techniques to improve segmentation accuracy.These approaches significantly increase the inference time.Other networks adopt more complex encoding and decoding modules to achieve end-to-end output,which requires substantial computation.Therefore,balancing the model’s size,accuracy,and computational complexity is essential.To address these challenges,we propose a lightweight asymmetric UNet architecture and a projection loss function.We utilize ResNet-3 layer blocks to enhance feature extraction efficiency in the encoding stage.In the decoding stage,we employ regular convolutions and skip connections to upscale the feature maps from the latent space to the original image size,balancing the model size and segmentation accuracy.In addition,we leverage the geometric features of the eye region and design a projection loss function to further improve the segmentation accuracy without adding any additional inference computational cost.We validate our approach on the OpenEDS2019 dataset for virtual reality and achieve state-of-the-art performance with 95.33%mean intersection over union(mIoU).Our model has only 0.63M parameters and 350 FPS,which are 68%and 200%of the state-of-the-art model RITNet,respectively.
文摘Medical image segmentation has become a cornerstone for many healthcare applications,allowing for the automated extraction of critical information from images such as Computed Tomography(CT)scans,Magnetic Resonance Imaging(MRIs),and X-rays.The introduction of U-Net in 2015 has significantly advanced segmentation capabilities,especially for small datasets commonly found in medical imaging.Since then,various modifications to the original U-Net architecture have been proposed to enhance segmentation accuracy and tackle challenges like class imbalance,data scarcity,and multi-modal image processing.This paper provides a detailed review and comparison of several U-Net-based architectures,focusing on their effectiveness in medical image segmentation tasks.We evaluate performance metrics such as Dice Similarity Coefficient(DSC)and Intersection over Union(IoU)across different U-Net variants including HmsU-Net,CrossU-Net,mResU-Net,and others.Our results indicate that architectural enhancements such as transformers,attention mechanisms,and residual connections improve segmentation performance across diverse medical imaging applications,including tumor detection,organ segmentation,and lesion identification.The study also identifies current challenges in the field,including data variability,limited dataset sizes,and issues with class imbalance.Based on these findings,the paper suggests potential future directions for improving the robustness and clinical applicability of U-Net-based models in medical image segmentation.
基金The researchers would like to thank the Deanship of Graduate Studies and Scientific Research at Qassim University for financial support(QU-APC-2025).
文摘Deep learning(DL),derived from the domain of Artificial Neural Networks(ANN),forms one of the most essential components of modern deep learning algorithms.DL segmentation models rely on layer-by-layer convolution-based feature representation,guided by forward and backward propagation.Acritical aspect of this process is the selection of an appropriate activation function(AF)to ensure robustmodel learning.However,existing activation functions often fail to effectively address the vanishing gradient problem or are complicated by the need for manual parameter tuning.Most current research on activation function design focuses on classification tasks using natural image datasets such asMNIST,CIFAR-10,and CIFAR-100.To address this gap,this study proposesMed-ReLU,a novel activation function specifically designed for medical image segmentation.Med-ReLU prevents deep learning models fromsuffering dead neurons or vanishing gradient issues.It is a hybrid activation function that combines the properties of ReLU and Softsign.For positive inputs,Med-ReLU adopts the linear behavior of ReLU to avoid vanishing gradients,while for negative inputs,it exhibits the Softsign’s polynomial convergence,ensuring robust training and avoiding inactive neurons across the training set.The training performance and segmentation accuracy ofMed-ReLU have been thoroughly evaluated,demonstrating stable learning behavior and resistance to overfitting.It consistently outperforms state-of-the-art activation functions inmedical image segmentation tasks.Designed as a parameter-free function,Med-ReLU is simple to implement in complex deep learning architectures,and its effectiveness spans various neural network models and anomaly detection scenarios.
基金supported by the STI2030-Major-Projects(No.2021ZD0200104)the National Natural Science Foundations of China under Grant 61771437.
文摘Neuronal soma segmentation plays a crucial role in neuroscience applications.However,the fine structure,such as boundaries,small-volume neuronal somata and fibers,are commonly present in cell images,which pose a challenge for accurate segmentation.In this paper,we propose a 3D semantic segmentation network for neuronal soma segmentation to address this issue.Using an encoding-decoding structure,we introduce a Multi-Scale feature extraction and Adaptive Weighting fusion module(MSAW)after each encoding block.The MSAW module can not only emphasize the fine structures via an upsampling strategy,but also provide pixel-wise weights to measure the importance of the multi-scale features.Additionally,a dynamic convolution instead of normal convolution is employed to better adapt the network to input data with different distributions.The proposed MSAW-based semantic segmentation network(MSAW-Net)was evaluated on three neuronal soma images from mouse brain and one neuronal soma image from macaque brain,demonstrating the efficiency of the proposed method.It achieved an F1 score of 91.8%on Fezf2-2A-CreER dataset,97.1%on LSL-H2B-GFP dataset,82.8%on Thy1-EGFP-Mline dataset,and 86.9%on macaque dataset,achieving improvements over the 3D U-Net model by 3.1%,3.3%,3.9%,and 2.3%,respectively.
文摘Thyroid nodules,a common disorder in the endocrine system,require accurate segmentation in ultrasound images for effective diagnosis and treatment.However,achieving precise segmentation remains a challenge due to various factors,including scattering noise,low contrast,and limited resolution in ultrasound images.Although existing segmentation models have made progress,they still suffer from several limitations,such as high error rates,low generalizability,overfitting,limited feature learning capability,etc.To address these challenges,this paper proposes a Multi-level Relation Transformer-based U-Net(MLRT-UNet)to improve thyroid nodule segmentation.The MLRTUNet leverages a novel Relation Transformer,which processes images at multiple scales,overcoming the limitations of traditional encoding methods.This transformer integrates both local and global features effectively through selfattention and cross-attention units,capturing intricate relationships within the data.The approach also introduces a Co-operative Transformer Fusion(CTF)module to combine multi-scale features from different encoding layers,enhancing the model’s ability to capture complex patterns in the data.Furthermore,the Relation Transformer block enhances long-distance dependencies during the decoding process,improving segmentation accuracy.Experimental results showthat the MLRT-UNet achieves high segmentation accuracy,reaching 98.2% on the Digital Database Thyroid Image(DDT)dataset,97.8% on the Thyroid Nodule 3493(TG3K)dataset,and 98.2% on the Thyroid Nodule3K(TN3K)dataset.These findings demonstrate that the proposed method significantly enhances the accuracy of thyroid nodule segmentation,addressing the limitations of existing models.
基金Princess Nourah Bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R826),Princess Nourah Bint Abdulrahman University,Riyadh,Saudi ArabiaNorthern Border University,Saudi Arabia,for supporting this work through project number(NBU-CRP-2025-2933).
文摘Brain tumor segmentation from Magnetic Resonance Imaging(MRI)supports neurologists and radiologists in analyzing tumors and developing personalized treatment plans,making it a crucial yet challenging task.Supervised models such as 3D U-Net perform well in this domain,but their accuracy significantly improves with appropriate preprocessing.This paper demonstrates the effectiveness of preprocessing in brain tumor segmentation by applying a pre-segmentation step based on the Generalized Gaussian Mixture Model(GGMM)to T1 contrastenhanced MRI scans from the BraTS 2020 dataset.The Expectation-Maximization(EM)algorithm is employed to estimate parameters for four tissue classes,generating a new pre-segmented channel that enhances the training and performance of the 3DU-Net model.The proposed GGMM+3D U-Net framework achieved a Dice coefficient of 0.88 for whole tumor segmentation,outperforming both the standard multiscale 3D U-Net(0.84)and MMU-Net(0.85).It also delivered higher Intersection over Union(IoU)scores compared to models trained without preprocessing or with simpler GMM-based segmentation.These results,supported by qualitative visualizations,suggest that GGMM-based preprocessing should be integrated into brain tumor segmentation pipelines to optimize performance.
基金funded by Umm Al-Qura University,Saudi Arabia under grant number:25UQU4300346GSSR08.
文摘This study presents an advanced method for post-mortem person identification using the segmentation of skeletal structures from chest X-ray images.The proposed approach employs the Attention U-Net architecture,enhanced with gated attention mechanisms,to refine segmentation by emphasizing spatially relevant anatomical features while suppressing irrelevant details.By isolating skeletal structures which remain stable over time compared to soft tissues,this method leverages bones as reliable biometric markers for identity verification.The model integrates custom-designed encoder and decoder blocks with attention gates,achieving high segmentation precision.To evaluate the impact of architectural choices,we conducted an ablation study comparing Attention U-Net with and without attentionmechanisms,alongside an analysis of data augmentation effects.Training and evaluation were performed on a curated chest X-ray dataset,with segmentation performance measured using Dice score,precision,and loss functions,achieving over 98% precision and 94% Dice score.The extracted bone structures were further processed to derive unique biometric patterns,enabling robust and privacy-preserving person identification.Our findings highlight the effectiveness of attentionmechanisms in improving segmentation accuracy and underscore the potential of chest bonebased biometrics in forensic and medical imaging.This work paves the way for integrating artificial intelligence into real-world forensic workflows,offering a non-invasive and reliable solution for post-mortem identification.
基金funded by Zhejiang Basic Public Welfare Research Project,grant number LZY24E060001supported by Guangzhou Development Zone Science and Technology(2021GH10,2020GH10,2023GH02)+1 种基金the University of Macao(MYRG2022-00271-FST)the Science and Technology Development Fund(FDCT)of Macao(0032/2022/A).
文摘With the continuous development of artificial intelligence and machine learning techniques,there have been effective methods supporting the work of dermatologist in the field of skin cancer detection.However,object significant challenges have been presented in accurately segmenting melanomas in dermoscopic images due to the objects that could interfere human observations,such as bubbles and scales.To address these challenges,we propose a dual U-Net network framework for skin melanoma segmentation.In our proposed architecture,we introduce several innovative components that aim to enhance the performance and capabilities of the traditional U-Net.First,we establish a novel framework that links two simplified U-Nets,enabling more comprehensive information exchange and feature integration throughout the network.Second,after cascading the second U-Net,we introduce a skip connection between the decoder and encoder networks,and incorporate a modified receptive field block(MRFB),which is designed to capture multi-scale spatial information.Third,to further enhance the feature representation capabilities,we add a multi-path convolution block attention module(MCBAM)to the first two layers of the first U-Net encoding,and integrate a new squeeze-and-excitation(SE)mechanism with residual connections in the second U-Net.To illustrate the performance of our proposed model,we conducted comprehensive experiments on widely recognized skin datasets.On the ISIC-2017 dataset,the IoU value of our proposed model increased from 0.6406 to 0.6819 and the Dice coefficient increased from 0.7625 to 0.8023.On the ISIC-2018 dataset,the IoU value of proposed model also improved from 0.7138 to 0.7709,while the Dice coefficient increased from 0.8285 to 0.8665.Furthermore,the generalization experiments conducted on the jaw cyst dataset from Quzhou People’s Hospital further verified the outstanding segmentation performance of the proposed model.These findings collectively affirm the potential of our approach as a valuable tool in supporting clinical decision-making in the field of skin cancer detection,as well as advancing research in medical image analysis.
