Search Results (1,493)

Accurate iris segmentation remains a fundamental challenge in iris biometric recognition and medical image analysis, particularly in challenging scenarios such as non-cooperative acquisition conditions involving variable illumination, partial occlusions, degraded image quality, and diverse unconstrained environments. Prevailing segmentation algorithms exhibit limited robustness when confronted with such challenges, and the disparity between near-infrared (NIR) and visible-light imaging modalities further compounds the complexity of achieving a robust segmentation outcome. To address these challenges, this paper introduces SwinIrisNet, a hybrid deep learning architecture that integrates Swin Transformer and convolutional neural network (CNN) branches within a U-Net framework for robust iris segmentation. The Swin Transformer branch leverages hierarchical window-based self-attention to capture global contextual dependencies, whereas the CNN branch extracts fine-grained local features essential for precise boundary delineation. A memory-efficient cross-attention fusion module combines these complementary feature representations, further enhanced by a Convolutional Block Attention Module (CBAM), Atrous Spatial Pyramid Pooling (ASPP), and attention-gated skip connections for multi-scale context aggregation. An extensive evaluation is conducted across four publicly available benchmark datasets, including UBIRIS.v2, IITD, CASIA-Thousand, and MMU.v1, encompassing both visible-light and NIR imaging environments. The proposed architecture yields F1 values of 0.9612–0.9672, Dice coefficients of 0.9489–0.9519, mIoU values of 0.9266–0.9450, precision values of 0.9565–0.9633, recall values of 0.9600–0.9672, and classification accuracies of 99.51–99.53%, with NICE1 error rates of 0.57–0.60% and NICE2 values of 1.82–2.24%, confirming pixel-level segmentation quality. Cross-database generalization experiments further demonstrate that SwinIrisNet learns transferable iris representations and generalizes effectively across heterogeneous imaging sources, with the strongest transfer occurring in the NIR-to-visible direction. A comparative analysis against existing algorithms demonstrates that the proposed architecture attains substantial performance improvements over several existing segmentation networks when evaluated on identical benchmark databases, surpassing them across the majority of qualitative and quantitative metrics while maintaining a marginally lower memory footprint. Full article

This survey presents a structured review of deep learning-based techniques for image data hiding, proposing a three-paradigm taxonomy organized by the method’s operational relationship to the carrier image. We classify existing methods into modification-based, synthesis-based, and logic-based approaches. In the modification-based tier, we trace the architectural progression from foundational Convolutional Neural Networks and Generative Adversarial Networks to high-capacity Invertible Neural Networks and Transformers, analyzing their distinct trade-offs between embedding capacity, imperceptibility, and robustness. In the synthesis-based tier, we examine how Diffusion Probabilistic Models and generative adversarial frameworks reframe data hiding as a carrier generation problem rather than a pixel editing task. This paradigm encompasses both generative steganography (where carriers are synthesized from scratch) and proactive watermarking (where provenance is embedded during AI content generation). In the logic-based tier, we review zero-watermarking and coverless steganography, where ownership is established through feature extraction and semantic mapping without modifying any image, a critical property for sensitive domains such as medical imaging. Finally, we identify four persistent infrastructure gaps: benchmarking fragmentation, narrow robustness evaluation, domain generalization failures, and computational infeasibility that prevent real-world deployment despite architectural progress, and we propose concrete research directions. Full article

Image segmentation is a fundamental step in computer vision and a cornerstone of robotic perception, serving as the foundation for interpreting data acquired from vision sensors, enabling robots to analyze complex visual environments, identify and localize objects, and support intelligent decision-making and autonomous control. It plays a critical role in applications such as autonomous navigation, robotic manipulation, medical robotics, agricultural robotics, autonomous vehicles, and human–robot interaction. Image segmentation has evolved from classical methods, which relied on handcrafted rules and mathematical models, to deep learning approaches that learn complex visual patterns directly from data. This evolution reflects advances in algorithms, computational power, and the theoretical foundations of mathematics and data science. Modern deep learning methods rely heavily on large, well-annotated datasets to train sophisticated neural networks. Yet, classical techniques remain valuable in certain scenarios, offering faster, reliable results without extensive computational requirements. Understanding the strengths and limitations of both approaches is key to selecting the right method. This paper surveys image segmentation techniques, comparing them in terms of accuracy, computational cost, and processing speed to guide informed method selection. Full article

Deep neural networks (DNNs) have achieved remarkable success in medical image classification, yet their performance remains sensitive to dataset size. Knowledge distillation (KD) alleviates this issue by transferring knowledge from a high-capacity teacher to a lightweight student. However, conventional KD relies on a static teacher, while adaptive teacher updating may improve performance on the student-learning data while reducing retention of knowledge acquired during teacher pretraining. To address these limitations, we propose an Anti-forgetting Adaptive Teacher-driven Knowledge Distillation framework (A²T-KD), which aims to balance teacher adaptation and pretraining-knowledge retention. The proposed framework integrates three modules: MITR for cross-epoch representation consistency, DSDO for prediction-space decoupling and class discriminability, and SGKD for feature- and logit-level knowledge transfer. Across nine medical imaging datasets, A²T-KD achieved higher mean values than the fixed-teacher Vanilla KD baseline in 30 of 36 dataset–metric comparisons. It also exhibited the lowest pretraining-set ACC degradation among the evaluated teacher-update baselines on all nine datasets, supporting the intended balance between teacher adaptation and pretraining-knowledge retention under the evaluated settings. Full article

Bone marrow lesion (BML) volume is an essential biomarker for understanding knee osteoarthritis (KOA). However, automatic BML segmentation remains challenging due to the irregular shapes and indistinct boundaries of these lesions in knee magnetic resonance images (MRI). To improve BML segmentation, this study investigated two established strategies in the specific context of BML segmentation: (1) integrating bone segmentation as an additional output channel in deep neural networks to facilitate BML segmentation, and (2) incorporating test-time augmentation (TTA) to reduce uncertainty during testing. The added bone segmentation channel provides auxiliary anatomical information that may facilitate BML localization. TTA was used to improve boundary alignment and reduce false positives by generating more robust predictions. Multiple State-of-the-Art deep neural networks for segmentation were employed as the baseline models to compare performance before and after implementing the proposed strategies. A 10-fold cross-validation was conducted on a dataset of knee MR scans from 300 participants. Segmentation performance was evaluated using the Dice similarity coefficient (DSC) for overlap accuracy and the 95% Hausdorff Distance (HD95) for boundary alignment. Paired t-tests were used to assess the significance of improvements from the proposed strategies. Both strategies produced improvements in segmentation performance, although the magnitude and statistical significance of the improvements varied across architectures. The DSC improved from 63.1% to 64.8% for Residual U-Net, 64.2% to 65.8% for Swin UNETR, 61.5% to 66.5% for Attention U-Net, and 66.6% to 69.0% for UNet++. These gains were accompanied by improvements in boundary accuracy and reductions in false positives, reflected in lower HD95 values. Comparison with additional medical image segmentation models under the same evaluation framework showed that the dual-channel UNet++ with TTA achieved the highest BML DSC of 69.0%, followed by U-Mamba at 68.6% and nnU-Net at 65.2%, while U-Net + InceptionResNet-v2 achieved 57.7%. These findings support the potential value of dual-channel and TTA strategies for automated BML analysis, while further validation on independent datasets is needed to assess their broader generalizability and clinical utility. Full article

Background/Objectives: To date, no systematic review has focused specifically on commercially available deep learning-based image reconstruction (DLIR) software for radiation dose optimization in pediatric radiology. The purpose of this article was to systematically review original clinical studies evaluating the performance of commercial DLIR software for radiation dose optimization in pediatric radiology and to assess their methodological quality. Methods: A literature search was conducted on 28 April 2026 using seven electronic databases. The review was registered with the Open Science Framework (Registration DOI: 10.17605/OSF.IO/3JDQS). Results: Thirteen papers met the selection criteria and were included in the review. These studies evaluated five commercial computed tomography (CT) DLIR software products. Clinically achievable radiation dose reductions ranged from 11.2% to 97.9% without compromising image quality, and in some cases, even exceeding that of the reference standard, suggesting further dose reduction potential. Excluding two studies that reported substantially lower dose reductions, the clinically achievable dose reduction range was 36.0–97.9%. Furthermore, more than three-quarters of the included articles reported dose reductions of at least 50% while maintaining image quality. Improvements in study methodology were evident among papers published from 2025 onward. Conclusions: Commercial CT DLIR software can achieve substantial radiation dose reductions in pediatric CT while maintaining image quality. However, further clinical studies are needed to evaluate a broader range of commercial DLIR software, including applications in positron emission tomography and X-ray imaging. Future studies should ideally include the full pediatric age range, prospectively collected and adequately sized datasets from underrepresented geographic regions, effective dose assessment, and clinically meaningful outcome measures such as diagnostic confidence/quality/accuracy. Full article

Although ConvNeXtV2 has shown promising performance in medical image classification, approaches relying primarily on final-stage features may underutilize low-level structural and complementary hierarchical information. To address this limitation, we propose an Adaptive Multi-Layer Feature Fusion Network (AMFF-Net) based on ConvNeXtV2-B for medical image classification. The proposed framework employs a Feature Alignment (FA) module to project multi-stage features into a unified representation space and an Adaptive Multi-Layer Feature Fusion (AMFF) module to compute a single input-dependent scalar weight for each stage and dynamically adjust the relative contributions of hierarchical features. An Efficient Channel Attention (ECA) module is subsequently incorporated to enhance the fused representation through lightweight channel-wise recalibration. AMFF-Net was evaluated on three medical image classification datasets: Kvasir-v2, HAM10000, and ChestXray14. Experimental results demonstrate that AMFF-Net consistently improves classification performance over the baseline ConvNeXtV2-B and achieves competitive performance compared with representative convolutional neural network (CNN)- and Transformer-based architectures, while incurring relatively modest additional computational overhead. Ablation results further support the contribution of FA, AMFF, and ECA to the overall performance of the proposed framework. Full article

Hyperparameter optimization is a critical determinant of the performance and generalizability of convolutional neural networks (CNNs), yet selecting optimal configurations remains a complex challenge due to the large, non-linear search space. While numerous optimization techniques have been explored in the literature, most comparative studies have relied on benchmark datasets, limiting insights into their practical effectiveness under real-world constraints. This study evaluates four widely used hyperparameter optimization methods (i.e., Grid Search, Optuna, Hyperopt, and DEAP) across five CNN architectures (i.e., AlexNet, GoogLeNet, LeNet, MobileNet, and ResNet18) under two distinct data regimes: the Modified National Institute of Standards and Technology (MNIST) benchmark and a custom SkinCancer image dataset. On MNIST, all configurations achieved strong classification performance, with test accuracies ranging from 95.45% to 97.61%, led by ResNet18 optimized via Grid Search (97.61%). In this simpler setting, DEAP provided the shortest optimization runtimes, while Optuna and Hyperopt offered an effective balance between performance and computational cost. In contrast, the SkinCancer dataset exhibited substantially greater variability in performance, highlighting the stronger influence of both model architecture and optimization strategy. ResNet18 achieved the highest overall test accuracy (81.50%) when optimized with Hyperopt. Although no single optimization method consistently outperformed all others, adaptive approaches frequently achieved competitive or superior predictive performance while requiring considerably less computation than exhaustive search. These findings demonstrate that hyperparameter optimization dynamics depend heavily on dataset complexity, where computational efficiency is the primary differentiator for simpler classification tasks, but optimization architecture selection becomes critical for navigating challenging medical imaging applications. Full article

In response to the challenge of balancing accuracy and generalization in medical image classification using convolutional neural networks (CNNs), this paper proposes ResKAN18, a hybrid structure that embeds the learnable spline function of the Kolmogorov–Arnold network (KAN) into the ResNet18 classification head for intelligent diagnosis of COVID-19 in chest X-ray images. ResKAN18 includes three variants: ResKAN18—Large (four layers of KAN, hidden-layer dimensions [256, 128, 64]), ResKAN18—Standard (four layers of KAN, hidden-layer dimensions [128, 64, 32]), and ResKAN18—Simple (three layers of KAN, hidden-layer dimensions [64, 32]), which can achieve a flexible balance between accuracy and efficiency with different depths of KAN. A systematic comparison has been conducted between four classic CNN baselines including ResNet18, VGG16, DenseNet121, ShuffleNetV2, and three ResKAN18 variants on a benchmark dataset containing 3880 chest X-rays (COVID-19, normal, viral pneumonia). The results have shown that ResKAN18—Large can achieve an accuracy of 98.80% on the independent test set, which is 1.21% higher than ResNet18 and 0.69% higher than DenseNet121—its parameter count is 13.97M, inference delay is 8.25 ms/image, and training–validation accuracy difference is only 1.50%. The accuracy and performance stability of the dataset under random partitioning conditions are superior to the other two variants and all classic CNN baselines. All ResKAN18 variants have achieved zero missed diagnoses for COVID-19, while ResNet18 has shown missed diagnoses (0.9944). Taking into account the trade-off between accuracy, generalization, and inference efficiency, ResKAN1—Large is recommended as the default configuration, while for edge deployment scenarios with severely limited resources, ResKAN18—Simple can provide a cost-effective alternative with an extremely low latency of 2.50 ms/image and only 3.1% parameter increment compared to ResNet18. Full article

Semantic segmentation is a critical task in computer vision aiming to identify and classify individual pixels in an image, with numerous applications, for example, in autonomous driving and medical image analysis. However, semantic segmentation can be highly challenging, particularly due to the need for large amounts of annotated data. Annotating images is a time-consuming and costly process, often requiring expert knowledge and significant effort; moreover, saving the annotated images could dramatically increase the storage space. In this paper, we propose a novel approach for semantic segmentation, requiring only rough information about the proportions of individual semantic classes, hereafter referred to as semantic proportions (SPs), rather than the necessity of ground-truth segmentation maps. This greatly simplifies the data annotation process and thus will significantly reduce the annotation time, cost and storage space, opening up new possibilities for semantic segmentation tasks where obtaining the full ground-truth segmentation maps may not be feasible or practical. Our proposed method of utilising semantic proportions can (i) further be utilised as a booster in the presence of ground-truth segmentation maps to gain performance without extra data and model complexity, and (ii) also be seen as a parameter-free plug-and-play module, which can be attached to existing deep neural networks designed for semantic segmentation. Extensive experimental results demonstrate the good performance of our method compared to benchmark methods that rely on ground-truth segmentation maps. Utilising semantic proportions suggested in this work offers a promising direction for future semantic segmentation research. Full article

Accurate morphological analysis of blood smears is vital for hematological diagnosis, yet manual examination is labor-intensive and subjective. While deep learning offers automation, its black-box nature and computational demands often hinder clinical trust and deployment. We propose XHIC-Net, an Explainable Hybrid Involution–Convolution Network designed for efficient and transparent cell classification. By integrating spatially adaptive involution operations with convolutional layers within a residual framework, XHIC-Net captures both contextual and fine-grained features efficiently. To enhance interpretability, a Grad-CAM-based explainable AI (XAI) module visualizes the cellular regions driving model predictions. The proposed framework was evaluated on a dataset comprising 12,879 microscopic blood smear images belonging to 12 morphological cell categories. Experimental results demonstrate that XHIC-Net achieves an overall accuracy of 98.88%, precision of 98.89%, recall of 98.87%, F1-score of 0.9887, and Cohen’s Kappa score of 0.9887. It outperformed established models, including DL models such as EfficientNetV2S, MobileNet family, DenseNet family, and VGG16, while using fewer parameters and requiring shorter training times. Furthermore, the XAI maps consistently highlighted biologically relevant structures, validating the model’s decision-making process. XHIC-Net is a strong, effective, and clear research model for automated hematology. With future clinical validation, it has the potential to be modified for point-of-care diagnostics in healthcare settings with limited resources. Full article

Background/Objectives: Paget’s disease of bone (PDB) and osteoporosis are chronic metabolic bone disorders characterized by disrupted bone remodeling and increased skeletal fragility; however, the underlying mechanism of PDB remains poorly understood. Artificial intelligence (AI) has emerged as a transformative tool in medical imaging, enabling automated feature extraction and improved diagnostic classification of skeletal disorders. This study aimed to investigate whether AI could distinguish subtle variations in bone morphology between PDB and osteoporotic bone. Methods: C57BL/6 mice femurs were scanned by µCT: 16 optineurin-knockout mice with a PDB phenotype (20–26 months), 25 genetically matched wild-type Aging mice (20–26 months), and 15 ovariectomized (OVX) mice with osteoporotic bone phenotype (4.5 months). Two AI algorithms were investigated: a machine learning (ML) model using 22 µCT-derived features trained with a Random Forest (RF) classifier, and a deep learning (DL) model using a 3D convolutional neural network (3D-CNN) trained on raw µCT images. Leave-one-out cross-validation was applied to evaluate model robustness. Results: Significant differences in volumetric, density, and morphological parameters of cortical and trabecular bone were observed between PDB and osteoporosis (p < 0.05). The RF algorithm achieved 90% accuracy in distinguishing PDB from both aging- and OVX-induced osteoporosis and provided feature importance rankings that improved model interpretability. The 3D-CNN achieved classification accuracies of 70% for PDB vs. OVX and 68% for PDB vs. aging, demonstrating the feasibility of an image-based DL approach. Conclusions: AI-based RF and 3D-CNN models demonstrated promising performance in differentiating PDB from osteoporosis using µCT-derived bone features. These findings suggest potential for using AI to assist with analyzing skeletal images in the diagnosis of metabolic bone disorders. Full article

Lung cancer arises from mutations in lung cells, disrupting their normal growth cycle and leading to uncontrolled cell division. These rapidly dividing cells lose function and fail to form healthy lung tissue. Several factors contribute to the difficulty of diagnosing and classifying lung nodules, including the high degree of morphological heterogeneity and overlapping characteristics between benign and malignant nodules. Recently, deep learning models have been used in computed tomography (CT)-based lung nodule diagnosis and have demonstrated diagnostic efficiency comparable to that of radiologists. This study introduces LungCNET, a high-performance multi-layer deep convolutional neural network trained on chest CT images to improve lung lesion classification efficiency and accuracy significantly. The data for the Lung Cancer convolutional neural network (LungCNET) were derived from the IQ-OTH/NCCD CT scan dataset (1097 images from 110 cases), split into training (767 images), validation (109) and a held-out test partition (221) that played no role in training or model selection. This dataset encompasses three diagnostic categories: benign, malignant, and normal lung tissues. LungCNET was evaluated against fine-tuned benchmark models that are both established and widely used, spanning architectures introduced between 2014 and 2024, including VGG16, ResNet50, InceptionV3, MobileNetV2, and YOLOv11. On the held-out test partition, LungCNET reached a macro-averaged F1-score of 95.19%, with VGG16 at 94.09% and InceptionV3 at 92.28%; these three models performed comparably, and the separation between them is small relative to the resolution of a test set of this size. LungCNET was, however, the only model to exceed 90% F1-score across all three diagnostic classes simultaneously, and recorded the highest F1-score on the benign class (91.0%), the smallest and most frequently misclassified category, where two of the six models failed entirely. These results support LungCNET as a candidate tool for lung cancer diagnosis, subject to validation on larger and independently sourced datasets. Full article

Background/Objectives: Class imbalance is one of the major challenges in automated skin lesion classification since the number of categories of malignant and clinically significant skin lesions is normally less than the benign ones. However, due to this imbalance, deep convolutional neural networks (CNNs) tend to overlook minority classes and fail to recognize them with an acceptable accuracy, which leads to a decrease in diagnostic reliability. A wide range of imbalance mitigation techniques has been suggested, but their effectiveness is found to differ significantly depending on CNN architecture, and detailed comparative studies of these techniques for a consistent experimental setup are still limited. Methods: This study proposes a comprehensive benchmarking framework that tests sixteen class imbalance mitigation methods by applying them to six pretrained CNN architectures—EfficientNet-B0, EfficientNet-B3, ResNet50, DenseNet121, InceptionV3 and MobileNetV2—on the official ISIC 2019 skin lesion dataset. The tested techniques are conventional resampling techniques, synthetic sample generation techniques, algorithm-level learning techniques, data augmentation techniques, and hybrid techniques. The dataset was partition into a separate training set and testing set, and stratified cross-validation was only conducted on the training set to ensure the study was fair and reproducible. Both models have been optimized with the same optimizer, learning rate, batch size, epochs and preprocessing pipeline. The performance of the models was evaluated by computing the accuracy, precision, recall and F1-score. Results: The experimental results show that the effect of class imbalance mitigation is very specific to the underlying CNN architecture. The traditional undersampling and oversampling methods yielded only moderate improvements, while feature space and hybrid methods yielded more consistent results. When coupled with EfficientNet-B3, Balanced MixUp improved the overall performance of the model by achieving an accuracy of 92.39%, an increase in precision of 93.3%, a recall of 91.36%, and an F1-score of 92.33%. However, some architectures such as ResNet50 performed better with iterative learning techniques, such as Cumulative Learning and Yielding Multi-Fold Training, which suggests that there is a diversity in how different network architectures react to imbalance mitigation methods. Conclusions: This paper highlights the importance of selecting appropriate technique–architecture combinations for addressing long-tailed data distributions in medical imaging. The proposed benchmarking framework provides valuable insights for developing robust and reliable deep learning systems for skin lesion classification and other medical imaging tasks affected by severe class imbalance. Full article

Background: Deep learning models, particularly convolutional neural networks (CNNs), have shown promising performance for pneumonia detection using chest X-ray images. However, the impact of preprocessing, architecture selection, data augmentation, and ensemble strategies has not been systematically evaluated. This study investigated how these factors affect model robustness and diagnostic performance. Methods: A public pediatric chest X-ray dataset was used to systematically evaluate pixel normalization methods, six CNN architectures, progressive data augmentation strategies for class imbalance, and both feature-level and decision-level ensemble approaches. Model performance was assessed by considering not only overall classification accuracy but also clinically relevant risk metrics, particularly false-negative rates. Results: Pixel normalization to the 0–1 range improved model convergence, while Xception and InceptionV3 achieved the best overall performance. Model-specific augmentation strategies were more effective than a fixed 1:1 class ratio for reducing false negatives. Feature-level ensembles tended to overfit, whereas decision-level ensembles provided more stable but only modest performance improvements. Conclusions: These findings demonstrate that reliable medical AI systems require systematic optimization of preprocessing techniques, model architecture, data augmentation strategies, and clinically meaningful evaluation metrics rather than maximizing a single performance indicator. The proposed framework provides practical guidelines for developing robust deep learning models for pneumonia diagnosis in clinical settings. Full article