International Journal of Imaging Systems and Technology最新文献

{"title":"Multi-Class Classification of Diabetic Retinopathy Diseases From Fundus Images Using Criss-Cross Attention Based Deep Residual Squeeze and Excitation Assisted Vision Transformer Model","authors":"Satti Mounika,&nbsp;V. RaviSankar","doi":"10.1002/ima.70199","DOIUrl":"https://doi.org/10.1002/ima.70199","url":null,"abstract":"<div>\u0000 \u0000 <p>The early detection of diabetic retinopathy (DR) illnesses improves diagnosis and lowers the risk of permanent blindness. As a result, screening for DR in fundus images is an important method used to diagnose diabetes and other eye diseases. However, detecting diseases manually requires a significant amount of time and work. Deep learning (DL) techniques have produced encouraging results in categorizing fundus images. Still, the multi-class DR disease remains a difficult task. Thus, the proposed framework adopted a novel Criss-Cross Attention-Based Squeeze-and-Excitation Assisted Vision Transformer (CCA-SE-ViT) model to classify DR from fundus images. Initially, the defective region of the retina is segmented using a novel dilated depth-wise separable convolutional U-Net model (dDSC-UNet). Then, using the segmented regions, the fundus images are classified into multiple classes as age-related macular degeneration (AMD), DR, glaucoma, cataracts, myopia, hypertension, normal, other abnormalities, and DR cases are classified into no DR, mild, moderate, severe, and proliferative DR, respectively. The retinal fundus images are obtained from publicly available datasets like OIA-ODIR and APTOS 2019. The proposed methodology for multi-class categorization of retinal illnesses in the OIA-ODIR dataset yielded 97.2% accuracy, 96.7% precision, 96.1% recall, 95.9% <i>F</i>1-score, and 96.4% specificity. The APTOS dataset was used for multi-class classification of DR illnesses, and the results were 99.68% accuracy, 99.08% precision, 99.31% recall, 99.19% <i>F</i>1-score, and 99.26% specificity. The results demonstrated that the proposed method accurately identifies DR using retinal fundus images.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145012702","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":"Multimodal Medical Image Fusion With UNet-Based Multi-Scale Transformer Networks","authors":"Qingwu Fu,&nbsp;Jianxuan Zhou,&nbsp;Jiao Du,&nbsp;Kai Lin,&nbsp;Bowen Zhong,&nbsp;Haoran Tang,&nbsp;Yiting Chen","doi":"10.1002/ima.70193","DOIUrl":"https://doi.org/10.1002/ima.70193","url":null,"abstract":"<div>\u0000 \u0000 <p>Multimodal medical image fusion can generate medical images that contain both functional metabolic information and structural tissue details, thereby providing doctors with more comprehensive information. Current deep learning-based methods often employ convolutional neural networks (CNNs) for feature extraction. However, CNNs exhibit limitations in capturing global contextual information compared to Transformers. Moreover, single-scale networks fail to exploit the complementary information between different scales, which limits their ability to fully capture rich image features and results in suboptimal fusion performance. To address these limitations, this paper proposes a multimodal medical image fusion method with UNet-based multi-scale Transformer network. First, we design a UNet-based encoder that incorporates a lightweight Transformer model, PVTv2, to extract multi-scale features from both MRI and SPECT images. To enhance the structural details of MRI images, we introduce the Edge-Guided Attention Module. Additionally, we propose an objective function that combines structural and pixel-level losses to optimize the proposed network. We perform both qualitative and quantitative experiments on mainstream datasets, and the results demonstrate that the proposed method outperforms several representative methods. In addition, we extend the proposed method to other biomedical functional and structural image fusion tasks, and the results show that the proposed method has good generalization capability.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145012628","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":"Dual Diversity and Pseudo-Label Correction Learning for Semi-Supervised Medical Image Segmentation","authors":"Guangxing Du,&nbsp;Rui Wu,&nbsp;Jinming Xu,&nbsp;Xiang Zeng,&nbsp;Shengwu Xiong","doi":"10.1002/ima.70194","DOIUrl":"https://doi.org/10.1002/ima.70194","url":null,"abstract":"<div>\u0000 \u0000 <p>Semi-supervised medical image segmentation has recently gained increasing research attention as it can reduce the need for large-scale annotated data. Current mainstream methods usually adopt two sub-networks and encourage the two models to make consistent predictions for the same segmentation task through consistency regularization. However, the scarcity of medical samples reduces the effectiveness of consistency constraints, and this problem may be further exacerbated by the influence of noisy pseudo-labels. In this work, we propose a novel co-training framework based on dual diversity and pseudo-label correction learning (DDPCL) to address these challenges. Specifically, firstly, we design a dual diversity learning strategy, in which data diversity fully mines the potential information of limited training samples through the CutMix operation, and feature diversity promotes the model to learn complementary feature representations by minimizing the similarity between the features extracted by the two sub-networks. Secondly, we propose a pseudo-label correction learning strategy, which regards the inconsistent region where the pseudo-labels predicted by the two sub-networks are different as potential bias regions, and guides the models to correct the bias in these regions. Extensive experiments on three public datasets (ACDC, LA and Pancreas-NIH datasets) validate that the proposed method outperforms the state-of-the-art semi-supervised medical image segmentation. The code is available at http://github.com/ddd0420/ddpcl.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144930054","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 Unified Framework for Reliability Analysis in Neuroimaging With Krippendorff's α","authors":"Mikkel C. Vinding","doi":"10.1002/ima.70192","DOIUrl":"https://doi.org/10.1002/ima.70192","url":null,"abstract":"<p>This paper proposes Krippendorff's <i>α</i> for reliability analysis of neuroimaging data. Reliability analysis quantifies the robustness of data and is crucial for ensuring consistent results across different analysis pipelines or methods. It measures the ratio between observed and expected agreement among raters using coincidence matrices. The paper explains how to calculate <i>α</i> and provides MATLAB code for implementation, along with examples of use on neuroimaging data. It includes a computationally efficient method for calculating <i>α</i> and a faster approximation method that maintains the logic of the exact test, making it suitable for large datasets typically found in neuroimaging. The uncertainty of the test statistic is estimated by bootstrapping. This work aims to simplify reliability analysis in neuroimaging, making it accessible for researchers.</p>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ima.70192","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144927485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Hybrid Swin Transformer and EfficientNetB2-Based Framework for Automated ROP Detection in Premature Infants","authors":"Ruoyao Cai,&nbsp;Jiaxuan Li,&nbsp;Fangzhou Wang,&nbsp;Xiaoya Li,&nbsp;Jiaming Guo,&nbsp;Jie Dai,&nbsp;Yingshan Shen","doi":"10.1002/ima.70191","DOIUrl":"https://doi.org/10.1002/ima.70191","url":null,"abstract":"<div>\u0000 \u0000 <p>Retinopathy of prematurity (ROP) is the leading cause of childhood blindness, and early and accurate detection is crucial for timely intervention and vision protection. To address the challenges of subtle lesion features and class imbalance in ROP images, this study proposes a hybrid deep learning model that integrates a generative-discriminative collaborative mechanism and a multi-module feature fusion strategy, which combines the local detail extraction of EfficientNet-B2 with the global context modeling of swin transformer to enhance the robustness of fine-grained lesion perception and classification. Experimental evaluation on the dataset shows that our model achieves 96.71% accuracy and 97.65% specificity, significantly outperforming mainstream baseline models. The performance improvement has been statistically validated (<i>p</i> &lt; 0.05). These results highlight the effectiveness of the model in addressing the challenges of ROP classification, providing a promising solution for intelligent assisted diagnosis, facilitating early disease warning, and promoting the application of artificial intelligence in ophthalmology.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144918840","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":"MHAOSL-Net: A Global Context-Aware Attention Free Transformer Network With Orthogonal SoftMax Layer to Detect Subtypes of Acute Lymphoblastic Leukemia","authors":"Rabul Saikia,&nbsp;Sakshi Gupta,&nbsp;Anupam Sarma,&nbsp;Ngangbam Herojit Singh,&nbsp;Deepak Gupta,&nbsp;Muhammad Attique Khan,&nbsp;Ashit Kumar Dutta,&nbsp;Salam Shuleenda Devi","doi":"10.1002/ima.70189","DOIUrl":"https://doi.org/10.1002/ima.70189","url":null,"abstract":"<div>\u0000 \u0000 <p>Recent advancements in Deep Learning (DL) have enabled the development of Computer-Aided Diagnosis (CAD) systems for detecting Acute Lymphoblastic Leukemia (ALL) and its subtypes. However, this field faces challenges, particularly due to limited annotated datasets and a lack of efficient modalities. To address these issues, we propose MHAOSL-net, a novel hybrid DL model specifically designed for the accurate classification of B-ALL, T-ALL, and normal cells in blood smear images. The key contributions lie in the integration of four primary components: (1) lightweight MobileNetV2 for backbone feature extraction, (2) a Global Context Information Convolutional Block Attention Module (GCI-CBAM) for refined local representation using contextual information, (3) an Attention-Free Transformer (AFT) that captures global dependencies replacing traditional self-attention, and (4) an Orthogonal SoftMax Layer (OSL) that improves class separability by enforcing orthogonality in the decision space. This unified architecture not only reduces computational overhead but also improves classification performance and generalizability. To the best of our knowledge, this is the first framework that combines an AFT with an OSL in the context of leukemia subtype detection. The performance analysis of the proposed 3-class classification scheme has been assessed on two novel datasets, namely <i>BBCI_B&amp;T_ALL_2024</i> and heterogeneous datasets. The experimental results show that the proposed scheme provides better performance, with 99.52% accuracy, 99.36% average precision, and 99.36% average F1-score on the <i>BBCI_B&amp;T_ALL_2024</i>. Similarly, it achieves better performance with 99.55% accuracy, 99.40% average precision, and 99.40% average F1-score on the heterogeneous dataset. The qualitative investigation using Gradient-weighted Class Activation Mapping (Grad-CAM) visualization also confirms the efficacy of the proposed model for detecting B-ALL, T-ALL, and normal cells. The comparative studies establish the superiority of the proposed scheme over other state-of-the-art approaches. These findings indicate that MHAOSL-Net offers a promising and efficient solution for reliable ALL subtype detection in clinical settings.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144910385","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 Lightweight Statistical Multi-Feature Adaptive Attention Network for Dermoscopic Image Segmentation","authors":"Weiye Cao,&nbsp;Kaiyan Zhu,&nbsp;Tong Liu,&nbsp;Jianhao Xu,&nbsp;Yue Liu,&nbsp;Weibo Song","doi":"10.1002/ima.70190","DOIUrl":"https://doi.org/10.1002/ima.70190","url":null,"abstract":"<div>\u0000 \u0000 <p>With the advent of Transformer architectures, the segmentation performance of dermoscopic images has been significantly enhanced. However, the substantial computational load associated with Transformers limits their feasibility for deployment on resource-constrained mobile devices. To address this challenge, we propose a Statistical Multi-feature Adaptive Attention Network (SFANet) that aims to achieve a balance between segmentation accuracy and computational efficiency. In SFANet, we propose a Multi-dilation Asymmetric Convolution Block (MDACB) and a Group Feature Mask Enhancement Component (GMEC). MDACB is composed of Multi-dilation Asymmetric Convolution (MDAC), a set of ultra-lightweight Statistical Multi-feature Adaptive Spatial Recalibration Attention (SASA) modules, Statistical Multi-feature Adaptive Channel Recalibration Attention (SACA) modules, and residual connections. MDAC efficiently captures a wider range of contextual information while maintaining a lightweight structure. SASA and SACA integrate multi-statistical features along spatial and channel dimensions, adaptively fusing mean, maximum, standard deviation, and energy via learnable weights. Convolution operations then model spatial dependencies and capture cross-channel interactions to generate attention weights, enabling precise feature recalibration in both dimensions. GMEC groups features from lower decoding layers and skip connections, and then merges them with the corresponding stage-generated masks, enabling efficient and accurate feature processing in the decoding layers while maintaining a low parameter count. Experiments on the ISIC2017, ISIC2018, and PH2 datasets demonstrate that SFANet achieves a mIoU of 80.15%, 81.12%, and 85.30%, with only 0.037 M parameters and 0.234 GFLOPs. Our code is publicly available at https://github.com/cwy1024/SFANet.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144910386","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":"Research on Multi-Objective Optimization of Medical Image Segmentation Based on Frequency Domain Decoupling and Dual Attention Mechanism","authors":"Xiaoling Zhou,&nbsp;Shili Wu,&nbsp;Yalu Qiao,&nbsp;Yongkun Guo,&nbsp;Chao Qian,&nbsp;Xinyou Zhang","doi":"10.1002/ima.70186","DOIUrl":"https://doi.org/10.1002/ima.70186","url":null,"abstract":"<div>\u0000 \u0000 <p>Medical image segmentation faces the challenge of balancing multiscale anatomical structure modeling and computational efficiency. To address this issue, this paper proposes a “Frequency-Attentive Multi-Hierarchical Network for Medical Image Segmentation” (FreqAtt-MultHier-Net), aiming to achieve synergistic optimization of accuracy, efficiency, and robustness. The core innovations of this paper include: A dual-frequency block (DFB), which decouples high-frequency (detail) and low-frequency (semantic) features through a learnable channel splitting mechanism, and enhances multiscale representations through cross-frequency interaction and dynamic calibration. A multiscale dual-attention fusion block (MSDAFB), which couples channel-spatial dual attention with multi-kernel convolutions to suppress background noise and strengthen local–global contextual fusion. A lightweight ConvMixer module that replaces Transformers with sublinear computational complexity to achieve efficient long-range dependency modeling. In tasks involving cell contour, cell nucleus, lung cancer, skin cancer, liver tumor segmentation and retinal vessel segmentation Task, our model achieves dice similarity coefficients (DSCs) of 95.64%, 92.74%, 83.63%, 85.96%, 85.86% and 84.26%, respectively, while reducing parameter count (25.48 M) and computational cost (5.84 G FLOPs) by 75.9%–84.9% compared to Transformer-based architectures. Ablation experiments validate the independent contributions of each module, with frequency-domain decoupling improving high-frequency detail retention by 18.8% and lightweight design reducing FLOPs by 78.3%. FreqAtt-MultHier-Net provides a high-precision, low-redundancy general solution for medical image segmentation, with potential for low-power clinical deployment. The code is available at the following URL: https://github.com/wu501-CPU/FreqAtt-MultHier-UNet.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144910384","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":"VMDUnet: Advancing Glioma Segmentation Integrating With Mamba and Dual Cross-Attention","authors":"Zhuo Chen,&nbsp;Yisong Wang,&nbsp;Fangfang Gou","doi":"10.1002/ima.70187","DOIUrl":"https://doi.org/10.1002/ima.70187","url":null,"abstract":"<div>\u0000 \u0000 <p>Gliomas are the most common type of primary brain tumor, characterized by their diffuse invasiveness and origin within the central nervous system. Manual identification and segmentation of tumor regions in MRI is a time-consuming and subjective process, and may negatively impact diagnostic accuracy because the heterogeneity and infiltrative pattern of glioma are complex. To address these problems, we propose an automated glioma segmentation approach named IADSG (Intelligent Assistant Diagnosis System for Glioma), based on our novel VMDUnet architecture. Our method incorporates Contrast Limited Adaptive Histogram Equalization (CLAHE) as a preprocessing step to enhance image contrast and quality. Moreover, we use data augmentation techniques to improve the generalization and adaptability to complex clinical images of the model. Crucially, the integration of a Mamba module and a dual cross-attention mechanism enables the model to effectively balance segmentation accuracy with computational efficiency. Experimental results show that our approach achieves a segmentation accuracy of 0.7769 DSC on the internal glioma dataset and 0.9117 DSC on the public BraTS dataset, outperforming existing segmentation methods on both benchmarks. This approach reduces the time and effort involved in manual segmentation, reduces the probabilities of misdiagnosis, and provides robust support for the diagnosis and treatment to be accurately conducted. Our code is available at https://github.com/CarioAo/VMDUnet.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144894408","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":"Deep Learning for Differentiating Active and Inactive Multiple Sclerosis Plaques: A Comparative Analysis of MRI-Based Classification Models","authors":"Mohammad Amin Shahram,&nbsp;Mostafa Robatjazi,&nbsp;Atefeh Rostami,&nbsp;Vahid Shahmaei,&nbsp;Ramin Shahrayini,&nbsp;Mohammad Salari","doi":"10.1002/ima.70188","DOIUrl":"https://doi.org/10.1002/ima.70188","url":null,"abstract":"<div>\u0000 \u0000 <p>Multiple sclerosis (MS) is a chronic inflammatory disease-causing neurological disability, particularly in young adults. Magnetic resonance imaging (MRI) is the most effective tool for detecting MS plaques, but contrast-enhanced imaging involves potential risks, including toxicity and increased imaging time. Previous methods for differentiating plaque types, such as texture analysis and manual feature extraction, face challenges such as limited datasets and poor generalizability. This study aims to develop and compare deep learning-based methods, specifically convolutional neural networks (CNNs), to classify MS lesion types using non-contrast MRI, aiming to improve clinical applicability and reduce reliance on contrast agents. This study involved 106 multiple sclerosis (MS) patients from two MRI centers. A total of 3410 lesions were analyzed, including 1408 active and 2002 inactive lesions. MRI images, including T1-weighted imaging with gadolinium contrast (T1 + Gd(, T1, Fluid-Attenuated Inversion Recovery (FLAIR), and T2 sequences, were acquired. The segmented lesions were converted into 2D slices and resampled to 128 × 128 pixels for deep learning input. Data augmentation and normalization were applied to improve model generalizability. A custom CNN model was developed and compared with four pre-trained models (ResNet50, VGG16, DenseNet121, and EfficientNetB0) using fivefold cross-validation to evaluate model performance. Performance metrics including accuracy, sensitivity, specificity, and AUC were used. The custom CNN achieved 90.15% accuracy and 94.67% AUC in FLAIR, outperforming pre-trained models. DenseNet121 showed competitive results with 88.23% accuracy and 92.86% AUC in FLAIR. Non-contrast sequences (T1, T2, and FLAIR) combined with deep learning provided promising results, reducing reliance on contrast agents. The custom CNN model excelled in classifying MS lesions across multiple MRI sequences, offering improved diagnostic accuracy and patient safety. Custom models for specialized datasets can enhance clinical outcomes, demonstrating the potential of deep learning in MS diagnosis. These findings suggest that deep learning models can be replaced with contrast agents in routine practice. Future research may explore combining CNNs with clinical features to enhance performance and interpretability.</p>\u0000 </div>","PeriodicalId":14027,"journal":{"name":"International Journal of Imaging Systems and Technology","volume":"35 5","pages":""},"PeriodicalIF":2.5,"publicationDate":"2025-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144892482","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}