Histopathology image analysis faces challenges due to magnification variability, limiting robust tumor categorization. Existing deep learning models prioritize accuracy but neglect explainability, ethical biases, and real-world deployment. This study proposes xMagNet, a hybrid Transformer-Convolutional Neural Network (CNN) framework that synergizes technical rigor, clinical transparency, and ethical fairness for multi-magnification breast cancer diagnostics. xMagNet integrates a hybrid encoder combining Vision Transformers (ViT) for global tissue modeling at low magnifications (4x - 10x) and Separable Dilation Convolutions (SDC) for localized nuclear texture extraction at high magnifications (20x - 40x). Magnification-Aware Gating (MAG) dynamically balances ViT and SDC features via temperature-scaled sigmoid activation. A multi-task decoder employs Thresholded Grad-CAM (top 10% gradients) for explainable decision-making and Point-wise Reformation Blocks (PRB) for boundary preservation. Federated learning (FL) with momentum-enhanced aggregation and Sinkhorn divergence regularization ensures scanner/stain-invariant training across six institutions (Hamamatsu/Leica, H&E/IHC). Uncertainty-quantified predictions (Monte Carlo dropout) and adversarial debiasing mitigate demographic leakage. xMagNet achieves 97.8% F1-score for tumor segmentation on Camelyon16 and 93% Gleason AUC on PANDA, with 96.5% pathologist concordance via Grad-CAM. At 40x magnification, it detects micro-metastases with 94% sensitivity (vs. UNet++’s 89% and ResUNet’s 91%). Computational efficiency includes sub-second inference (0.42 sec/slide) and 2.3 x faster convergence than HoVer-Net. Ethical auditing reveals 3% fairness gaps ( ) and 73% domain shift reduction (MMD: 0.12 vs. FedAvg’s 0.45), validated on 15,000 whole-slide images (WSIs) from TCGA-BRCA, Camelyon16, and PANDA datasets. xMagNet bridges critical gaps in multi-magnification histopathology by harmonizing technical robustness (MAG fusion, bounded gradients) with clinical utility (HER2+/ER+ subtyping, Gleason grading) and ethical scalability. By achieving high accuracy, rapid inference, and equitable deployment, it advances AI-driven diagnostics toward trustworthy, deployable systems for breast, prostate, and metastatic cancer imaging. Code available at: xMagNet .

Federated Class-Incremental Learning faces critical challenges including catastrophic forgetting, semantic drift, and privacy risks under non-IID data distributions. To address these, we propose FedCapD, a novel framework that unifies unsupervised task boundary detection via Bayesian nonparametric modeling, hierarchical semantic distillation through capsule alignment and GNN-based class propagation, secure gradient communication using homomorphic encryption and differential privacy, and diffusion-based generative replay for memory-efficient adaptation. By integrating structural reasoning with privacy-preserving learning, FedCapD achieves state-of-the-art performance across four diverse datasets-CheXpert, MIMIC-CXR-JPG, BraTS2021, and PHM2012-in terms of semantic consistency, encrypted distillation fidelity, cold-start accuracy, and utility efficiency under privacy constraints. The framework eliminates reliance on real-data storage and supports scalable, lifelong learning in regulated, resource-constrained environments such as healthcare AI and edge computing. Code is available at FCIL .

Robust and reliable medical diagnosis using artificial intelligence is crucial, yet real-world clinical environments present significant challenges due to dynamic covariate shifts affecting multi-modal data (images, text, tabular). Existing methods, including the single-modal robust classifier LaDiNE, often fail under these complex, multi-modal shifts, lacking mechanisms for cross-modal invariance, dynamic modality fusion, and fine-grained uncertainty attribution. To address this gap, we propose DyMoLaDiNE (Dynamic Multi-Modal Latent Diffusion Nested-Ensembles), a framework designed for reliable medical diagnosis under dynamic multi-modal covariate shifts. DyMoLaDiNE introduces four key innovations: (1) a Cross-Modal Invariant Feature Extractor leveraging multi-modal Vision Transformers and contrastive learning to derive robust latent representations, (2) a Dynamic Modality Weighting Mechanism that adaptively adjusts modality contributions based on instance-specific reliability scores, (3) a Robust Multi-Modal Diffusion Ensemble utilizing conditional diffusion models conditioned on multi-modal inputs and reliability scores for flexible, calibrated density estimation, and (4) Modality-Attributed Uncertainty Quantification to decompose predictive uncertainty by input source. Extensive evaluations on diverse datasets (MedMD&RadMD, MultiCaRe, PadChest, TCIA RE-MIND, BRaTS, Camelyon16, PANDA) demonstrate that DyMoLaDiNE significantly outperforms (p 0.005) state-of-the-art methods (LDM, CMCL, CGMCL, CIIM, DTTL, FFL, ALDM, LaDiNE) in terms of classification accuracy, robustness under dynamic perturbations, confidence calibration (ECE), and precise uncertainty quantification (CPIW, CNPV), while providing superior modality attribution fidelity. Ablation studies confirm the necessity of each component. DyMoLaDiNE represents a significant advancement in trustworthy, robust multi-modal medical AI. Code supporting this study DyMoLaDiNE .

This paper addresses the challenge of enhancing medical imaging analysis on edge devices while maintaining patient privacy and security. In this paper, we present a novel federated autoencoder model, U-NeTrans, which prioritizes security and privacy and is designed for medical image reconstruction on edge devices. U-NeTrans uses random masking to increase training complexity while maintaining manageability by using partial data. Data secrecy is ensured by the encoder processing visible patches and the decoder using encoded data to reassemble the original image. U-NeTrans improves the representation of high-order features in medical images by combining auxiliary reconstruction tasks and contrastive loss. This allows for precise analysis while maintaining patient privacy. The proposed method has wide ramifications for chest X-ray analysis and other medical imaging applications and offers the potential to improve healthcare device capabilities at the edge significantly. Comparative experimental results with benchmark datasets highlight the effectiveness of U-NeTrans compared to state-of-the-art approaches for edge-based medical image analysis while maintaining security and privacy. Accuracy, precision, sensitivity, specificity, and AUROC are measured across multiple scales and are shown to total 98.97%, 98.68%, 98.73%, and 99.19%, respectively.

ABSTRACT In the context of Cloud and Fog computing settings, recent developments in deep learning techniques show great potential for changing several fields, including healthcare. In this study, we make a contribution to this changing field by proposing an enhanced deep learning‐based strategy for classifying chest X‐ray images, using pre‐trained models such as RetinaNet, EfficientNet and Faster‐R‐CNN, which we use through transfer learning. Our strategy outperforms single models and traditional techniques by leveraging critical data gleaned from multiple models, demonstrating the ability of deep learning to improve diagnostic precision. Our approach presents a novel dual‐check system in the context of worries about security, privacy and trust in Cloud and Fog‐based Smart Systems. In this case, a decision support system uses chest X‐ray images to make an initial diagnosis that is then confirmed by a medical practitioner. This cooperative strategy not only reduces diagnostic errors that come from machine and human sources but also emphasises how crucial it is to incorporate AI‐driven solutions into safe and reliable healthcare ecosystems. Our approach raises the bar for the quality of patient care and healthcare outcomes by overcoming the drawbacks of traditional diagnostic methods that depend on the subjective opinions of physicians. Our work brings out how deep learning might transform clinical diagnostics by distinguishing inflammatory regions in chest X‐ray images. Research is needed to fully grasp the transformative potential of deep learning in medical image processing, especially as the healthcare industry continues to embrace AI‐driven solutions. Further research endeavours have to dig into tactics like broadening the scope of datasets, executing data augmentation methodologies and incorporating bespoken features to augment the elasticity and effectiveness of AI‐driven diagnostic systems.

ABSTRACT Chronic kidney disease (CKD) is a major global health concern caused mostly by high blood pressure and glucose levels. Detecting CKD early is critical for reducing its negative consequences since it can lead to increased mortality rates. With CKD's rising incidence expected to make it the fifth biggest cause of death by 2040, rapid advances in diagnostic approaches are required. This study presents the Reciprocal Domain Adaptation Network (RDAN) as a potential approach to the various issues of CKD diagnosis. RDAN is a neural network model that will help to traverse the complexity of CKD diagnosis by smoothly combining diverse data sets. RDAN consists of two critical units at its foundation: Mutual Model Adaptation (MMA) and Domain Model Learning. The MMA unit uses a powerful Global and Local Pyramid Pooling technique to extract rich features from a variety of data domains. Meanwhile, the DML unit uses semi‐supervised domain‐independent features combined with MMA features to improve representation learning. RDAN includes a reciprocal regularizer to promote cross‐domain knowledge transfer, maximising feature representation for accurate CKD identification. An analysis of RDAN's performance on a variety of real‐world datasets showed remarkable results in terms of accuracy (96.94%), precision (98.81%), recall (98.73%), F1‐Score (98.88%), and area under the curve (AUC—99.35%). These results highlight the unmatched expertise of RDAN in managing data bias, domain changes, and privacy issues related to CKD diagnosis. Beyond statistical measures, RDAN's implications promise revolutionary breakthroughs in early CKD identification and subsequent therapeutic therapies. RDAN stands out as a groundbreaking method for diagnosing CKD. It delivers exceptional accuracy and can be seamlessly applied in various clinical environments.

Electrocardiogram (ECG) arrhythmia identification is critical for early cardiovascular disease diagnosis and monitoring in Internet of Things (IoT) industry. Still, it is difficult due to complicated waveforms, individual variability, and the requirement for real-time analysis on resource-limited equipment. Traditional approaches sometimes fail to detect complicated spatial-temporal correlations in ECG data, limiting their efficiency in identifying arrhythmias. Furthermore, deploying these models in tinyML contexts, such as edge and IoT devices limited by large computational and memory needs, emphasizes the importance of lightweight, accurate models for real-time applications. Our suggested solution consists of three main components: SigNet, DualGCNN, and FusionGCNN. SigNet uses Separable Convolution layers to effectively extract local spatial features, making it ideal for IoT-based healthcare deployment. DualGCNN combines dual Graph Convolutional layers with spatial attention, allowing the model to capture local and global dependencies for better classification of arrhythmia. FusionGCNN combines the capabilities of GCN and SigNet with an effective feature fusion technique to improve feature representation while remaining computationally economical. Ablation tests show that FusionGCNN improves performance considerably, with greater accuracy (0.9641), lower training error (0.0004), and a higher F1 Score (0.9645) across a variety of ECG patterns. FusionGCNN, with its low training error, high stability, and computational economy, is well-suited to tinyML requirements, allowing implementation on edge and IoT devices for scalable, real-time ECG monitoring in healthcare.

The Hybrid Parallel Fuzzy CNN (HP-FCNN) is a ground-breaking method for medical image analysis that combines the interpretive capacity of fuzzy logic with the capabilities of a convolutional neural network (CNN). This novel combination tackles problems related to brain image processing, reducing problems such as noise and hazy borders that are common in Magnetic Resonance Imaging (MRI). Unlike other CNN models, HP-FCNN combines fine-grained fuzzy representations with crisp CNN features, improving interpretability by displaying hidden layers. This insight into activation patterns facilitates comprehension of the decision-making processes necessary for the diagnosis of brain diseases. HP-FCNN outperforms other pretrained models (ResNet, DenseNet, VGG, and EfficientNet) on measures such as the confusion matrix and AUC-ROC, according to comparative assessments. Furthermore, the addition of Adaptive Class Activation Mapping (AD-CAM) enhances HPFCNN by identifying salient features during backpropagation and bolstering the network's capacity to enhance brain illness diagnosis and treatment planning. Our methodology, incorporating AD-CAM, yielded compelling results with a 96.86 F1-Score, 96.41 AUC, and 96.81 Accuracy, showcasing the effectiveness of our approach in achieving high-performance metrics in brain MRI analysis. With a 15% increase in accuracy, a 10% increase in sensitivity, and a 12% decrease in false positives, HP-FCNN outperforms its predecessors. These impressive advancements represent a quantifiable breakthrough in the capabilities of medical image processing technology; they are more than just anecdotal evidence.

In the realm of machine vision, the convolutional neural network (CNN) is a frequently used and significant deep learning method. It is challenging to comprehend how predictions are formed since the inner workings of CNNs are sometimes seen as a black box. As a result, there has been an increase in interest among AI experts in creating AI systems that are easier to understand. Many strategies have shown promise in improving the interpretability of CNNs, including Class Activation Map (CAM), Grad-CAM, LIME, and other CAM-based approaches. These methods do, however, have certain drawbacks, such as architectural constraints or the requirement for gradient computations. We provide a simple framework termed Adaptive Learning based CAM (Adaptive-CAM) to take advantage of the connection between activation maps and network predictions. This framework includes temporarily masking particular feature maps. According to the Average Drop-Coherence-Complexity (ADCC) metrics, our method outperformed Score-CAM and another CAM-based activation map strategy in Residual Network-based models. With the exception of the VGG16 model, which witnessed a 1.94% decline in performance, the performance improvement spans from 3.78% to 7.72%. Additionally, Adaptive-CAM generates saliency maps that are on par with CAM-based methods and around 153 times superior to other CAM-based methods.