International Journal of Imaging Systems and Technology最新文献

{"title":"Lightweight Skin Lesion Segmentation Network With Multi-Scale Feature Fusion Interaction","authors":"Xiaofen Jia,&nbsp;Wenyang Wang,&nbsp;Zhenhuan Liang,&nbsp;Baiting Zhao,&nbsp;Mei Zhang,&nbsp;Cong Wang","doi":"10.1002/ima.70176","DOIUrl":"https://doi.org/10.1002/ima.70176","url":null,"abstract":"<div>\u0000 \u0000 <p>The existing segmentation algorithms have many problems, such as a large number of parameters, a complicated calculation process, and difficulty in accurately segmenting skin lesion areas with hair interference, blurred edges, and unclear lesion features. We propose a lightweight skin lesions segmentation network (LSLS-Net) to address the above problems. In the part of encoded feature extraction, we extract multi-scale features through different sizes of convolution kernels to capture rich detailed features of the skin lesion area; then we use a feature fusion enhancement module to enhance the extracted features. That is, we design a lightweight feature extraction module that extracts global features, an edge feature enhancement module that enhances edge features, and a feature fusion attention module that fuses and enhances global features and edge features. At the same time, the obtained different feature information is interfused with the unenhanced features to obtain more abundant features. Experimental results on two public datasets, ISIC-2018 and PH2, show that compared with current mainstream medical image segmentation algorithms UNet, AttentionUNet, UNet++, DoubleU-Net, CACDU-Net, EIU-Net, and HmsU-Net, the proposed algorithm not only obtains excellent performance in the number of parameters and computational complexity but also has a clear outline and continuous edge for the segmentation of skin lesions, which has a better segmentation effect. Additionally, experiments on the PH2 dataset further show that LSLS-Net possesses strong generalization capabilities.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144740489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ELK-BiONet: Efficient Large-Kernel Convolution Enhanced Recurrent Bidirectional Connection Encoding and Decoding Structure for Skin Lesions Segmentation","authors":"Jingjing Ma,&nbsp;Zhanxu Liu,&nbsp;Zhiqiang Guo,&nbsp;Ping Wang","doi":"10.1002/ima.70172","DOIUrl":"https://doi.org/10.1002/ima.70172","url":null,"abstract":"<div>\u0000 \u0000 <p>The size and shape of skin lesions often exhibit significant variability, and enabling deep learning networks to adapt to this variability is crucial for improving the segmentation performance of such lesions. The encoder-decoder architecture has become one of the most commonly used structures for semantic segmentation in deep learning models. However, when the convolution-based UNet network is applied to skin lesion segmentation, several issues remain. (1) Traditional small-kernel convolutions have a limited receptive field, which makes it difficult to adapt to the varying sizes and shapes of skin lesions. (2) The conventional U-Net architecture experiences a substantial increase in parameter count as the network depth grows. (3) Although the U-Net decoder utilizes feature information from the encoder, the features extracted by the decoder are not fully leveraged. To address the above challenges in U-Net for skin lesion segmentation tasks, we propose an efficient large-kernel convolution enhanced recurrent bidirectional connection encoding and decoding structure for skin lesions segmentation (ELK-BiONet). The main innovations of this method are as follows: (1) We propose a large-kernel convolution method that balances large and small receptive fields while maintaining a relatively low parameter count. (2) The network extracts feature information in a recurrent manner, allowing the construction of deeper network architectures while keeping the overall parameter count nearly constant. (3) By employing bidirectional connections, the features extracted by the decoder are fully utilized in the encoder, thereby enhancing the segmentation performance of the network. We evaluated our method on skin lesion segmentation tasks, and the results demonstrate that our ELK-BiONet significantly outperforms other segmentation methods.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144725487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Metaheuristic-Based Conductivity Distribution Optimization Method for Accurate Imaging in Electrical Impedance Tomography","authors":"Yanyan Shi,&nbsp;Yan Cui,&nbsp;Meng Wang,&nbsp;Zhenkun Liu,&nbsp;Feng Fu","doi":"10.1002/ima.70175","DOIUrl":"https://doi.org/10.1002/ima.70175","url":null,"abstract":"<div>\u0000 \u0000 <p>In the medical application of electrical impedance tomography (EIT), image reconstruction of conductivity distribution is essential for diagnosis of physiological or pathological changes. In this study, a metaheuristic-based conductivity distribution optimization method is proposed for accurate reconstruction. To test the performance, simulation work is conducted and different models are reconstructed. Images reconstructed by the Newton–Raphson method, Tikhonov method, and genetic algorithm have been adopted for comparison. The effect of noise on the proposed method is also investigated. In addition to simulation, a phantom experiment is designed to further testify to the effectiveness of the proposed method. The results show that the proposed method outperforms other comparative methods in conductivity distribution imaging. The proposed method gives a more precise reconstruction of the inclusion, with a notably clearer background. Meanwhile, the proposed method is more robust to noise. It offers an effective alternative for conductivity distribution reconstruction in the application of EIT.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144714805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards Efficient Brain Tumor Segmentation via a Transformer-Driven 3D U-Net","authors":"Runlin Chen,&nbsp;Huixuan Luo,&nbsp;Yanming Ren,&nbsp;Wenjie Liu,&nbsp;Wenyao Cui","doi":"10.1002/ima.70158","DOIUrl":"https://doi.org/10.1002/ima.70158","url":null,"abstract":"<div>\u0000 \u0000 <p>Accurate brain tumor segmentation is critical for clinical diagnosis and treatment. The rapid development of deep neural networks (DNNs) in computer vision offers an automated solution for segmentation tasks. However, convolutional neural networks (CNNs) cannot model long-range dependencies, hindering their perception of global information on tumors. Moreover, vision Transformers (ViTs) require extensive annotated data for optimal segmentation performance, leading to high computational costs and overfitting on small datasets. To address these challenges, we propose TDU-Net, an efficient and accurate brain tumor segmentation scheme using Transformer-driven 3D U-Net. In TDU-Net, improved inverted residual bottlenecks with large kernels are employed in both downsampling and upsampling blocks, optimizing memory efficiency while maintaining global semantic richness in 3D multimodal tumor data. Inspired by ViT, fewer activation functions and normalization layers are used in downsampling and upsampling blocks. GELU activation, group normalization, and larger convolution kernels are employed to improve the global perception and segmentation capability on small datasets. Additionally, orthogonal regularization is introduced during training to mitigate overfitting and enhance generalizability. Experimental results demonstrate that TDU-Net achieves superior brain tumor segmentation accuracy with fewer model parameters, thereby improving generalizability and reducing performance degradation due to overfitting.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modified Transformer-Based Pixel Segmentation for Breast Tumor Detection","authors":"Kamakshi Rautela,&nbsp;Dinesh Kumar,&nbsp;Vijay Kumar","doi":"10.1002/ima.70166","DOIUrl":"https://doi.org/10.1002/ima.70166","url":null,"abstract":"<div>\u0000 \u0000 <p>This study introduces a novel hybrid deep learning model that combines residual convolutional networks and a multilayer perceptron (MLP)-based transformer for precise breast lesion segmentation and classification using mammogram images. Initially, mammograms undergo preprocessing involving thresholding and Gabor-based pixel segmentation to extract informative patches. The proposed model leverages deep features extracted via convolutional neural networks, which are subsequently processed through self-attention and cross-attention mechanisms in a modified transformer architecture to capture both local and global dependencies for classification. The approach is rigorously evaluated on the publicly available INbreast dataset, achieving classification accuracies of 98.17% for a three-class (normal, benign, malignant) scenario and 96.74% for a more detailed five-class classification. The model demonstrates strong capabilities in differentiating subtle variations between malignant and benign tissues. These promising results suggest significant potential for practical clinical implementation, assisting radiologists by providing highly accurate diagnostic insights. Notably, this approach contributes substantially to automated breast cancer diagnostics, highlighting the efficacy of integrating convolutional neural network features with transformer architectures for improved segmentation and classification outcomes.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144681247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CMFR-Net: Cross Multi-Scale Features Refinement Network for Medical Image Segmentation","authors":"Huimin Guo,&nbsp;Yonglai Zhang,&nbsp;Hualing Li,&nbsp;Gaizhen Liu,&nbsp;Jiaxin Huo","doi":"10.1002/ima.70167","DOIUrl":"https://doi.org/10.1002/ima.70167","url":null,"abstract":"<div>\u0000 \u0000 <p>The automation of medical image segmentation can assist doctors in quickly and accurately extracting lesion regions, reducing their workload in clinical analysis, improving diagnostic efficiency, and aiding in the early diagnosis and analysis of diseases. However, medical images are susceptible to noise, and variations in the position, size, and shape of organs and tissue structures across different patients pose significant challenges in achieving accurate segmentation. In this paper, we propose the Cross Multi-scale Features Refinement Network (CMFR-Net), which introduces the cross features enhancement (CFE) module, the boundary refinement (BR) module, and the global context features guidance (GCFG) module to extract multi-scale spatial information and boundary details of the target region, capture long-range feature dependencies, and improve segmentation performance. The CFE module captures local feature information from target regions at different scales, the BR module alleviates boundary blurring issues during segmentation, and the GCFG module strengthens the model's ability to capture global features and spatial positional information. Experiments conducted on three public datasets and one private dataset demonstrate the effectiveness of the proposed CMFR-Net. The Dice coefficients of CMFR-Net on the four datasets reached 87.35%, 87.65%, 97.52%, and 88.38%, respectively.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144673120","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Real-Time, Multi-Task Mobile Application for Automatic Bleeding and Non-Bleeding Frame Analysis in Video Capsule Endoscopy Using an Ensemble of Faster R-CNN and LinkNet","authors":"Divyansh Nautiyal,&nbsp;Manas Dhir,&nbsp;Tanisha Singh,&nbsp;Anushka Saini,&nbsp;Palak Handa","doi":"10.1002/ima.70171","DOIUrl":"https://doi.org/10.1002/ima.70171","url":null,"abstract":"<div>\u0000 \u0000 <p>Real-time, multi-task mobile application for automatic bleeding and non-bleeding frame analysis in video capsule endoscopy (VCE) frames is critical for early diagnosis but is currently underexplored. This study presents a mobile application using Flutter that can automatically classify VCE frames as bleeding and non-bleeding, and further identify and segment bleeding areas in real time. The application utilizes an ensemble deep learning model that integrates Faster Region-based Convolutional Neural Network (R-CNN) for frame-level classification and LinkNet for pixel-level segmentation. Faster R-CNN first detects and classifies VCE frames as bleeding or non-bleeding, and subsequently, LinkNet segments the bleeding regions within the frames identified as bleeding. Both models were trained and validated using the publicly available WCEBleedGen dataset. To evaluate the effectiveness of the proposed ensemble, a comparative analysis was conducted with existing studies and state-of-the-art (SOTA) models in the field. For detection, the performance of Faster R-CNN was compared with two You Only Look Once (YOLO) variants: YOLOv5 and YOLOv12. For segmentation, LinkNet was compared with SegNet and UNet. Evaluation metrics included mean Average Precision at 0.5 ([email protected]), Dice coefficient, and Eigen class activation maps. The mobile application achieved an average inference time of 2.88 s per frame and 23.33 s for a batch of 10 frames. Overall, the ensemble model attained a [email protected] of 0.92 and a Dice coefficient of 0.96, outperforming existing studies. For SOTA models, Faster R-CNN outperformed YOLO variants by achieving a 25% higher [email protected], and LinkNet achieved a 26% higher Dice coefficient than SegNet and 5% higher than UNet on the validation dataset and achieved more focused Eigen maps for different bleeding areas. This study represents the first attempt to develop a real-time, multi-task mobile application for VCE bleeding analysis. The application is open-source and freely available at https://github.com/misahub2023/VCE-BleedGen-Application, supporting accessibility, reproducibility, and future research in this field.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144673122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Study of Retinal Vessel Segmentation Algorithm Based on Receptive Field Expansion and Feature Refinement","authors":"Xinghua Wang,&nbsp;Jiawen Cao,&nbsp;Runxin Meng,&nbsp;Xiaolong Liu,&nbsp;Jie Wang,&nbsp;Yuting Tang,&nbsp;Ruijin Sun","doi":"10.1002/ima.70164","DOIUrl":"https://doi.org/10.1002/ima.70164","url":null,"abstract":"<div>\u0000 \u0000 <p>Missing blood vessels, fracturing blood vessels, and mistaking nonvascular features for blood vessels are major problems in retinal vessel segmentation tasks. This paper suggests an enhanced model that incorporates the Inception module and attention mechanism, based on the U-Net network topology, to solve these problems. In order to get richer scale information and enhance the model's recognition of vascular details, the encoder portion of the model first employs convolution kernels of varying sizes to collect multilevel characteristics of the picture. Second, to enhance feature processing between codecs and highlight significant features, an attention module is integrated into skip connections to extract spatial location information and interchannel interactions. This information is then coupled with residual connections. Finally, in the decoding stage, a residual attention module was constructed to extract vascular features and improve processing speed. On the DRIVE standard fundus image dataset, the proposed algorithm demonstrates significant performance enhancements compared to the conventional U-Net baseline. Specifically, it achieves absolute improvements of 1.94% in sensitivity, 1.07% in Jaccard index, 0.75% in Dice correlation coefficient, and 0.74% in Matthews correlation coefficient. Compared with other algorithms, it also has certain advantages and can effectively perform retinal vessel segmentation.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MaxGlaViT: A Novel Lightweight Vision Transformer-Based Approach for Early Diagnosis of Glaucoma Stages From Fundus Images","authors":"Mustafa Yurdakul,&nbsp;Kübra Uyar,&nbsp;Şakir Taşdemir","doi":"10.1002/ima.70159","DOIUrl":"https://doi.org/10.1002/ima.70159","url":null,"abstract":"<div>\u0000 \u0000 <p>Glaucoma is a prevalent eye disease that often progresses without symptoms and can lead to permanent vision loss if not detected early. The limited number of specialists and overcrowded clinics worldwide make it difficult to detect the disease at an early stage. Deep learning-based computer-aided diagnosis (CAD) systems are a solution to this problem, enabling faster and more accurate diagnosis. In this study, we proposed MaxGlaViT, a novel Vision Transformer model based on MaxViT to diagnose different stages of glaucoma. The architecture of the model is constructed in three steps: (i) the Multi Axis Vision Transformer (MaxViT) structure is scaled in terms of the number of blocks and channels, (ii) low-level feature extraction is improved by integrating the attention mechanism into the stem block, and (iii) high-level feature extraction is improved by using the modern convolutional structure. The MaxGlaViT model was tested on the HDV1 fundus image data set and compared to a total of 80 deep learning models. The results show that the MaxGlaViT model, which contains effective block structures, outperforms previous literature methods in terms of both parameter efficiency and classification accuracy. The model performs particularly high success in detecting the early stages of glaucoma. MaxGlaViT is an effective solution for multistage diagnosis of glaucoma with low computational cost and high accuracy. In this respect, it can be considered as a candidate for a scalable and reliable CAD system applicable in clinical settings.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657719","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Differentiating Sporadic Colonic Hamartoma From Adenomas in Narrow Band Imaging Using a Novel AI Network: Attention Based Multi-Scale CNN (AM-Net)","authors":"Aditi Jain,&nbsp;Saugata Sinha,&nbsp;Bhargava Chinni,&nbsp;Srijan Mazumdar","doi":"10.1002/ima.70168","DOIUrl":"https://doi.org/10.1002/ima.70168","url":null,"abstract":"<div>\u0000 \u0000 <p>There are no existing protocols for optical diagnosis of Sporadic colonic hamartomas, which are benign polyps, using the narrow-band imaging (NBI). Efficient detection of hamartoma polyps is difficult due to the similar appearances in NBI with other polyp types. Differentiating hamartoma from adenomatous is necessary for efficient utilization of “diagnose and leave” or “resect and discard” strategies during colonoscopy procedure. To address the above challenge, we conducted a study where suitably trained AI algorithms were employed for automatic differentiation of hamartoma and adenomatous polyps. An Attention based Multi-scale CNN (AM-Net), that integrates a Multi-scale Residual Network (MRN) with a parallel attention module (PAM) was introduced in this study. The Multi-scale Residual Network (MRN) structure enables the model to capture local multi-scale features while the attention module identifies “where to focus” and “what to focus on” through channel and spatial dimensional attention. To the best of our knowledge, AM-Net is the first AI-based model designed to differentiate colonic hamartomas from adenomatous polyps using NBI colonoscopy videos. In this study the performance of AM-Net was evaluated using a real-life colonoscopy polyp video comprising 1706 NBI polyp frames collected from 45 patients at a tertiary care hospital. The dataset includes 761 frames of hamartoma polyps and 945 frames of adenomatous polyps. The results demonstrated that efficient differentiation between hamartoma and adenomatous polyps is possible using a suitably designed and trained AI network. The proposed AM-Net achieved an accuracy of 86.97%, precision of 82.84%, F1-score of 87.75%, and AUC of 0.95, outperforming existing state-of-the-art CNN architectures and attention mechanisms across all metrics by effectively capturing structural details such as polyp mucosal patterns, textures, and boundaries, showcasing its ability to substantially enhance the accurate classification of hamartoma polyps.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 4","pages":""},"PeriodicalIF":3.0,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}