Jul 2026· International Conference on Computer Communications and Networks· pp. 1-6· 0 citations· 25 references
Abstract
Deep learning in medical imaging is severely hindered by the "domain gap," where significant imaging discrepancies exist even among images of the same modality and anatomical structure due to heterogeneous scanning devices and diverse clinical environments. While major medical centers possess high-fidelity data, resource-constrained clinics often struggle with legacy hardware and severe data scarcity. This imbalance leads to a significant collapse in diagnostic performance and poor model generalization when deployed in low-resource settings. To address these challenges, we propose a novel few-shot Federated Low-Rank Adaptation (Fed-LoRA) framework for cross-domain generative feedback. By leveraging the lightweight nature of LoRA, our method enables the efficient transmission of global pathological priors from a high-resource Hub to decentralized Edge nodes with minimal communication overhead. Crucially, this mechanism facilitates cross-domain data generation while ensuring privacy by precluding the transmission of raw medical data. This allows local models to synthesize high-fidelity abnormal exemplars precisely tailored to specific institutional styles. Extensive experiments demonstrate that our synthetic images significantly enhance the performance of downstream anomaly detection tasks. This work provides a scalable and secure solution for mitigating data scarcity and achieving effective cross-domain adaptation in decentralized healthcare systems.
In medical imaging, generative models are increasingly deployed to synthesize realistic data and augment limited datasets. Unfortunately, while beneficial for privacy-preserving data sharing, these synthesized images can be repurposed for malicious intents, threatening public health through the creation of Medical Deepfakes. To address this threat, we introduce the K-Space Signature (KSS), a novel forensic framework that isolates hardware and generative traces within the spectral domain. By shifting analysis to the frequency domain, the KSS suppresses macroscopic anatomical variance by subtracting an empirical global anatomical prior computed in the Logarithmic Power Spectral Density (Log-PSD) space. To effectively process these globally distributed spectral artifacts without the local spatial bias inherent to Convolutional Neural Networks, we pair the KSS representation with a novel 3D MLP-Mixer architecture equipped with an ArcFace metric-learning head. Extensive experiments on multi-center 3D MRI datasets demonstrate that this combined approach achieves exceptional detection performance, exceeding 0.99 Accuracy and ROC-AUC on multi-generator synthetic datasets. Furthermore, the framework exhibits robust zero-shot generalization, maintaining strong discriminative power (up to 0.93 Accuracy) on independent datasets acquired from entirely unseen scanners. To ensure full reproducibility, the complete source code and pre-trained models will be made publicly available upon acceptance.
Riccardo Raciti, Francesco Guarnera, F. Rundo et al.· 0 citations
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
Medical image datasets often suffer from limited training samples and severe class imbalance. This scarcity hinders deep learning models from learning representative features and leads to degraded performance on minority classes. In this work, we adapt the Unconditional Training at Lower Resolutions (UTLO) framework to long-tailed brain MRI data and conduct systematic evaluations on a brain tumor dataset. UTLO enables knowledge sharing between frequent and rare classes by training the generator unconditionally at lower resolutions to capture class-agnostic structures, while employing conditional generation at higher resolutions to refine class-specific details. Meanwhile, the discriminator is optimized with a multi-objective loss across low- and high-resolution stages to improve robustness and alleviate mode collapse under limited data. We evaluate the generated images using Frechet Inception Distance (FID), Kernel Inception Distance (KID), Inception Score (IS), and their few-shot variants, i.e., FID-FS (FID on few-shot/rare classes) and KID-FS (KID on few-shot/rare classes). Qualitative visualization is also provided to assess perceptual realism and diversity. Experimental results demonstrate that UTLO achieves stable generation performance across categories and yields noticeable improvements on rare classes.
Lung ultrasound (LUS), valued for its portability and AI-driven analysis, has become an essential di agnostic tool in emergency medicine, critical care, and the screening of infectious diseases. However, the scarcity of annotated data and limited expert availability remain major barriers to large-scale deployment. In this study, we pro pose a Similarity-Guided Multi-Source Few-Shot Learning (SM-FSL) paradigm that shifts the focus from domain alignment to domain selection, enabling robust feature learning from multiple semantically relevant domains under limited data conditions. Based on this paradigm, we develop the Global-Local Multi-Source Domain Network (GLMD-Net) for LUS analysis in data-scarce scenarios, which first employs the proposed Average Nearest Neighbor Set Distance (ANNSD) to prioritize relevant source domains for pre-training. A similarity-guided optimization mechanism then harmonizes multi-source gradient updates, while a local feature enhancement module and a self-supervised auxiliary task improve robustness against noise and artifacts. Finally, a few-shot adaptation strategy fine-tunes only the classification head for efficient and stable knowledge transfer. Extensive experiments on two public COVID-19 ultrasound datasets demonstrate that our method surpasses state-of-the-art approaches, validating its effectiveness in enhancing generalization and robustness under few-shot settings.
Enguang Wang, Juxiang Wu, Shuo Gao et al.· IEEE journal of biomedical a...· 0 citations
Extensive experiments across three medical imaging benchmarks, two brain tumor classification datasets (SARTAJ, Br35H) and dental radiography analysis demonstrate that CENet variants achieve state-of-the-art efficiency-accuracy trade-off.
Amina Benabid, Kangjie Cheng, Yun-Feng Liu et al.· Journal of King Saud Univers...· 0 citations
Adapting deep learning models to profound clinical heterogeneity typically relies on parameter-efficient fine-tuning (PEFT) to avoid the severe overfitting associated with full end-to-end network updates. Although PEFT successfully navigates limited data scenarios, it inherently forces the training of a separate, isolated adapter for every specific diagnostic task. Consolidating these isolated adapters into a single generalist network risks negative transfer, as optimization gradients from conflicting visual domains interfere. To address this, we propose MoPET, a mixture-of-experts (MoE) method that uses a learned sparse router to direct each input through a small subset of low-rank PEFT experts injected into a frozen foundation model, sharing capacity across datasets while limiting cross-domain gradient conflict. Through selected evaluations on the MedMNIST benchmark, we first establish that PEFT outperforms full network updates, improving average accuracy from 86.50% to 88.97%. We then show that a single MoPET model consolidates four heterogeneous datasets into one network, improving average accuracy over the best isolated PEFT adapters (93.46% versus 92.83%). Finally, we show that co-training with auxiliary datasets improves accuracy on data-constrained clinical targets, raising average target accuracy over the strongest isolated adapter from 81.58% to 83.58%. Our source code is publicly available at https://github.com/sdoerrich97/mopet.
Sebastian Doerrich, Daniel Würtinger, Francesco Di Salvo 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.
