Jun 2026· IEEE International Conference on Healthcare Informatics· pp. 1550-1559· 0 citations· 25 references
Computer Science
TL;DR
This work proposes an analytic federated learning framework for multi-label medical image classification under task heterogeneity that consistently outperforms the state-of-the-art federated multi-label method FedMLP and is backbone-agnostic and generalizes across ResNet, VGG, and EfficientNet encoders without any hyperparameter adjustment.
Abstract
Federated learning (FL) enables multiple clinical institutions to collaboratively train a shared disease classifier without centralizing patient data. In practice, however, each institution annotates only the pathologies within its area of expertise, so the federation operates under task heterogeneity: each client holds labels for a strict subset of the target disease categories while the remaining classes are entirely unobserved at that site. Existing gradient-based FL methods fail under this setting because they require hundreds of communication rounds to converge and because missing class labels introduce systematic false-negative bias that the model cannot correct without a principled mechanism. We propose an analytic federated learning framework for multi-label medical image classification under task heterogeneity. The proposed method replaces iterative gradient optimization with three closed-form operations: a balanced label projection that neutralizes class-imbalance bias by normalizing positive and negative contributions to equal total mass; a per-class absolute aggregation law that independently assembles the optimal ridge-regression classifier for each disease category from the sufficient statistics uploaded by its annotating clients; and an optional analytic pseudo-label refinement round that propagates missingclass knowledge from a confidence-filtered teacher classifier to non-annotating clients. The entire procedure requires at most two communication rounds, irrespective of the degree of task heterogeneity or the number of participating clients. Experiments on ChestXray14 under four progressively severe missing-class configurations demonstrate that the proposed method consistently outperforms the state-of-the-art federated multi-label method FedMLP by up to 18.44 BACC points and 13.24 AUC points, while reducing the communication. Further experiments across different backbone architectures confirm that the proposed method is backbone-agnostic and generalizes across ResNet, VGG, and EfficientNet encoders without any hyperparameter adjustment.
STPFL builds an EMA-based aggregated teacher to accumulate historical global knowledge and provide consistent guidance and improves global accuracy and F1-score, while personalized models improve by 4–25% and 4–32%, respectively.
R. Semwal, Imlimaong Aier, P. Varadwaj· Intelligent Data Analysis· 0 citations
This work proposes a label-free selection criterion built on SUDO, a framework for evaluating clinical AI systems without ground-truth annotations, and shows that AURCC can be used to rank a variety of vision-language models on chest X-ray classification across three inter-hospital shift scenarios, under zero-shot and MLP-probe regimes.
Juan Iñaki Larrea, L. Mansilla, Enzo Ferrante· 0 citations
Federated learning enables medical-imaging models to be trained across hospitals, and privacy law, most explicitly the GDPR ``right to be forgotten'', turns removing a hospital's, a class's, or a patient's influence from such a model into a federated unlearning problem. This need is most acute in medicine, where patients withdraw consent and hospitals leave collaborations. Yet nearly all unlearning evidence comes from natural images, whose heterogeneity and task structure differ sharply from clinical data, so it is unclear whether existing methods transfer, and no shared protocol covers clinical data. We present Lethe, a benchmark for federated unlearning in medical imaging. It evaluates twelve methods across eight task families, from classification and segmentation to denoising, cross-modality synthesis, and vision-language question answering, at three forgetting granularities and against a retrained gold standard on utility, privacy, and cost. The central result is that what separates methods is the difficulty of the forgetting request, not the method itself. The easy removals that dominate the literature leave the methods that preserve utility indistinguishable, while only hard ones separate them. More striking, on the many medical tasks that generalize across sites, forgetting a client barely changes task performance, leaving residual membership as the signal that must be erased.
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
Diagnostic decision making often relies on a sequence of pathology tests that bridge patient symptoms and final disease diagnosis. Existing clinical decision-support systems typically focus on predicting single diseases and do not explicitly recommend sets of intermediate tests or model dependencies among them. In this paper, we formulate pathology test recommendation as a multi-label classification problem where each case is associated with multiple, interdependent tests. We propose an AI-based framework that applies classifier chains with logistic regression, decision trees, random forests, and their ensemble to capture label dependencies between tests. Experiments on an expert-curated dataset from a private pathology laboratory show that classifier-chain models outperform their independent counterparts, improving F1-score and reducing Hamming loss while maintaining high accuracy across common and rare tests. To enhance trust and transparency, we integrate SHAP-based explainable AI, providing symptom-level attributions that align with established clinical reasoning in most cases. The results demonstrate that classifier chains combined with SHAP offer an effective and interpretable approach for multi-label pathology test recommendation, with potential to support clinicians in selecting appropriate diagnostic tests at an early stage.
A novel class-incremental continual learning model for a one-shot FL paradigm, in which each task introduces new classes, clients observe heterogeneous and evolving class distributions, and communication with the server occurs only once, substantially mitigates catastrophic forgetting while consistently enhancing recognition of newly introduced classes.
Pedro H. Barros, Omid Orang, Giulia Zanon de Castro et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
