This work explores CXAI use cases in multi-label classification by training two DNNs, VGG16 and ResNet50, on the 20 most annotated labels in the MS-COCO dataset and demonstrates the potential of CXAI to enhance the understanding of model generalizability and to diagnose bias instigated by the dataset.
Abstract
Deep Neural Networks (DNNs) deployed in high-risk domains, such as healthcare and autonomous driving, must be not only accurate but also understandable to ensure user trust. In real-world computer vision tasks, these models often operate on complex images containing background noise and are heavily annotated. To make such models explainable, Concept-based Explainable AI (CXAI) methods need to be assessed for their applicability and problem-solving capacity. In this work, we explore CXAI use cases in multi-label classification by training two DNNs, VGG16 and ResNet50, on the 20 most annotated labels in the MS-COCO dataset (Microsoft Common Objects in Context). We apply two CXAI methods, CRP (Concept Relevance Propagation) and CRAFT (Concept Recursive Activation FacTorization), to generate concept-level explanations and investigate the overall evaluations. Our analysis reveals three key findings: (1) CXAI highlights learning weaknesses in DNNs, (2) higher concept distinctiveness reduces label and concept confusion, and (3) environmental concepts expose dataset-induced biases. Our results demonstrate the potential of CXAI to enhance the understanding of model generalizability and to diagnose bias instigated by the dataset.
Skin cancer remains a global health threat with increasing incidence rates. Accurate and early classification of skin lesions into benign or malignant types is essential for timely treatment and prevention of severe outcomes. In this paper, we present a comprehensive deep learning-based framework that leverages three benchmark datasets—PH2, ISIC (Benign vs Malignant), and HAM10000—using transfer learning and ensemble techniques. Pre-trained models including VGG16, ResNet50, and EfficientNetB4 were fine-tuned on each dataset, and majority voting was employed to combine predictions. The Gradient-weighted Class Activation Mapping (Grad-CAM) was also used to improve visual explainability. The findings demonstrate a notable increase in classification accuracy, surpassing current techniques and reaching over 98% accuracy on certain datasets. This study highlights the impact of hybrid architectures and explainable AI in advancing the state of skin cancer diagnosis systems.
A. D. Hayder, J. Saeed· passer of basic and applied...· 0 citations
The precise identification of thoracic disorders using chest X-rays (CXRs) is essential for favorable clinical diagnosis. However, it continues to be difficult because of inter-class similarities and overlapping symptoms. This study introduces a hybrid MobileNetV2–Vision Transformer (ViT) framework for multi-label classification of CXR images into 14 disease categories on NIH CXR14 dataset. MobileNetV2 serves as a lightweight feature extractor to obtain distinctive spatial representations that are tokenized and enhanced with positional encodings. The encoded features were processed via transformer layers, wherein self-attention mechanisms captured long-range interdependence and global contextual relationships. An output layer based on the sigmoid function facilitated the concurrent prediction of various illness labels. The novelty of the proposed work lies in the PSO-tuned hybrid architecture, which adaptively optimizes key hyperparameters of the MobileNetV2–ViT framework to achieve improved accuracy, faster convergence, and enhanced computational efficiency compared to conventional approaches. This implementation uses an economical, GPU-accelerated instance tailored for inference workloads. It is equipped with a single NVIDIA A10G GPU with 24 GB of dedicated GPU memory, and the proposed work is carried out on AWS SageMaker. The experimental assessment employing five-fold cross-validation showed strong performance, with an average accuracy of 94.6%, precision of 88.1%, recall of 85.9%, specificity of 94.17%, and F1-score of 91.1%. Grad-CAM representations increased interpretability by identifying disease-relevant areas. The results underscore the efficacy of the proposed hybrid technique as a dependable instrument for computer-aided diagnosis of thoracic illnesses.
R. Raj, Pavan M. P. Kumar, K. N. Manjunath et al.· Scientific Reports· 0 citations
Alzheimer’s disease (AD) remains a critical global health challenge for which early diagnosis is essential for effective intervention. However, AD detection is a challenging and complex task due to the scarcity of labeled clinical data, class imbalance, and lack of prediction interpretability in healthcare systems. Although current machine learning frameworks are promising, they are frequently based on fully supervised environments and lack transparency, which limits their ability to be adopted in real-world clinical scenarios where labeled data are often scarce. This study proposes a new lightweight, semi-supervised, SHAP-enhanced CatBoost (LSSSC) framework that covers the challenges related to insufficient label data, interpretability of prediction models, and performance in a single learning paradigm. The proposed solution provides two major innovations: (i) a hybrid semi-supervised learning algorithm that incorporates the use of confidence-aware pseudo-labeling with cost-sensitive learning, allowing the effective use of the unlabeled data and alleviating the problem of class imbalance, and (ii) an intrinsic SHAP-based feature selection mechanism that assists not only in increasing interpretability but also in minimizing the dimensionality of the feature variables. The suggested approach is tested on a real AD dataset of 2149 patients and 34 diverse clinical, demographic, and lifestyle characteristics. Experimental findings show that LSSSC outperforms traditional supervised and ensemble frameworks, with an accuracy of 95.35%, macro-F1 score of 0.9490, and AUC of 0.9513. Moreover, the combination of SHAP and LIME provides global and local interpretability, permitting a meaningful understanding of the model’s predictions. These contributions make LSSSC a strong, scalable, and interpretable framework for early AD prediction and promise much with respect to its application to real-world clinical decision-support systems.
Abdallah El Chakik, B. Nakhal, Ghalia Nassreddine· The Scientist· 0 citations
Reliable automated Chest X-ray (CXR) analysis is critical for clinical decision support, yet deep learning models often suffer from overconfidence, limited interpretability, and poor out-of-distribution robustness. We propose a framework towards trustworthy AI for multi-class CXR classification integrating a hybrid CNN–Transformer with bi-cross attention, uncertainty quantification, and neuro-symbolic reasoning. The architecture fuses EfficientNet-B3 and ViT-B/16 features via bi-cross attention, aligning local and global representations. Predictive trust is assessed using Expected Calibration Error (ECE), while class-conditional Mahalanobis distance enables out-of-distribution (OOD) detection. Explainability is enhanced by constraining Grad-CAM++ heatmaps within lung masks. A neuro-symbolic reasoning (NSR) module translates neural activations into human-readable diagnostic statements grounded in radiological priors. Experiments on a four-class CXR dataset achieve 96% accuracy, macro F1-score of 0.97, and strong calibration (ECE = 0.0328). For OOD detection, the method achieves AUROC = 0.97 on out-of-domain images. These results demonstrate that combining bi-cross attention, anatomy-aware explainability, and neuro-symbolic logic yields a transparent and robust framework – a step towards clinically trustworthy AI for medical imaging.
Y. Didi, Ahlam Walha, Ali Wali· International Conference on...· 0 citations
Deep learning has shown strong potential in medical image analysis, but most existing methods rely on large-scale annotations and a closed-world assumption that rarely holds in clinical practice. Although Generalized Category Discovery (GCD) has advanced rapidly on natural images, it remains underexplored in medical imaging. To address this issue, we propose MedXplore, a unified framework for reliable and unbiased medical GCD, optimizing from both perceptual and decision levels. Specifically, at the perceptual level, taking a frequency domain perspective, Frequency-SNR Adaptive Attention and Consistency (FAAC) performs learnable full-spectrum filtering and global-local energy contrast activation to not only highlight local abnormal signals relative to the global context, but also provide reliable semantic anchors for patch consistency learning. At the decision level, Adaptive Cosine-Angular Margin (ACAM) adjusts angular margins using semantic difficulty and feature confidence to balance intra-class compactness and inter-class separability. Together, the two modules improve lesion-sensitive representation learning and mitigate old-class bias. Experiments on multiple benchmarks show an average \textbf{8.5\%} gain in \textit{All} accuracy over the strongest competing methods. On Kvasir, MedXplore reduces false-old errors from 14.50\% to 0.80\%, demonstrating strong robustness under severe old-new ambiguity.
OBJECTIVE
Existing prototype learning methods for open-set recognition (OSR) often use a fixed number of prototypes to represent each class, which struggle to model the inherent intra class variations widely existing in practical scenarios. This limitation is particularly pronounced in medical image applications, where high intra-class heterogeneity makes accurate OSR challenging. To address this issue, we propose a novel framework called Adaptive Multi-Prototype Network with Pretrained Swin Transformer (PSW-AMPN) for OSR on medical images. Specifically, a sparse gated attention module is devised to compute attention scores based on prototype-sample relations, thereby adaptively suppressing redundant sub-prototypes and selectively activating discriminative ones for each class in an end to-end optimization process. By jointly optimizing classification loss and the regularization for open space risk based on multiple prototypes, PSW-AMPN effectively captures complex intra-class structures and enhances class discriminability. Furthermore, PSW-AMPN uses a Pretrained Swin Transformer and a lightweight projector as feature extractor to effectively capture both local and global features. Extensive experiments demonstrate that our approach significantly outperforms existing baselines and achieves state-of-the-art performance on multiple medical image OSR tasks.
Xingyu Cai, Haiyan Yang, Jiayi Chen et al.· IEEE transactions on bio-med...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
