• Proposes SN-WGAIN-GP, a deep learning framework for imputing SHM data under random and continuous missing scenarios. • Integrates different strategies for stable and efficient adversarial training of the proposed model. • Employs Monte Carlo dropout, Gaussian noise, and critic-guided pooling methods for uncertainty-aware multiple imputation. • Validated on LUMO and Canton Tower datasets across varying noise levels and temporal resolutions. • Achieves RMSE < 0.10 and PCC > 0.95 under severe missingness, outperforming other benchmark imputation models. Reliable data acquisition is fundamental to the integrity of Structural Health Monitoring (SHM) systems and their integration within digital twin environments. However, data loss caused by sensor malfunction, communication failure, or environmental interference remains a persistent challenge, undermining model fidelity and structural condition assessment accuracy, especially during extreme loading events when continuous monitoring is most critical for digital twin applications. To address this, we propose SN-WGAIN-GP, a novel Spectrally Normalised Wasserstein Generative Adversarial Imputation Network with Gradient Penalty, designed for robust multi-modal time-series data imputation in SHM applications. The framework enhances the adversarial learning process through three key mechanisms: spectral normalisation to enforce Lipschitz continuity and stabilise training; wasserstein distance with gradient penalty to improve convergence and reduce mode collapse; and a two-time-scale update rule to balance generator-critic dynamics and accelerate optimisation. SN-WGAIN-GP operates in both single and multiple-imputation modes. In the latter, it leverages Monte Carlo (MC) dropout, Gaussian noise, and multiple latent noise vectors to generate diverse imputed datasets, followed by critic-weighted softmax pooling for aggregation and Rubin’s rule for quantifying predictive uncertainty. Comprehensive evaluation on two benchmark datasets, the LUMO experimental steel mast and the Canton Tower full-scale instrumented structure, demonstrates that SN-WGAIN-GP consistently outperforms existing GAN-based baseline models across a wide range of missing-data rates, noise levels, and temporal resolutions. Validation through operational modal analysis confirms accurate natural frequency extraction, with imputed responses showing excellent agreement with true responses in both time-domain and frequency spectra. SN-WGAIN-GP maintains correlation coefficients above 0.95 and RMSE below 0.10 even under 90% data loss, while achieving computational efficiency compatible with real-time digital twin deployment. The framework provides a scalable, uncertainty-aware and physically interpretable solution for multi-sensor SHM data imputation under partial observability, advancing the reliability of data-driven digital twins for predictive maintenance, resilience assessment, and intelligent infrastructure management.

Accurate segmentation of stroke lesions in diffusion-weighted imaging (DWI) is crucial for clinical decision-making. However, automated infarct segmentation remains challenging due to variable infarct sizes and locations, and it is labor-intensive, requiring expert manual annotations for training. We propose a zero-shot framework to eliminate the need for manual segmentation labels by leveraging weak supervision from class activation maps (CAMs) to guide segmentation using MedSAM2, a foundation model for 3D medical image segmentation. By extracting attention maps from a fine-tuned ResNet on DWI scans labeled with stroke etiology (cause) and combining them with intensity information, we identify key regions and generate bounding-box prompts for MedSAM2. Our method achieves a Dice score of 54.2 ± 5.3% without any manual segmentation labels or tuning of the MedSAM2 model, demonstrating its potential as a scalable solution for reliable pseudo-label generation.

Medical image synthesis addresses healthcare data scarcity by generating realistic samples for clinical support systems, AI training, and research. However, the field faces challenges due to the complexity of imaging data with its diverse modalities, characteristics, and disease variations. To produce high-quality images, medical image synthesis typically relies on conditional generation, where labels and annotations serve as essential conditions that provide critical guidance signals during the generation process to control desired semantics and fidelity. However, in the medical domain, labels are often inaccessible due to the high cost of annotation, requirements for clinical expertise, as well as ethical concerns. To address this critical challenge, we propose RepMedGAN, a novel self-supervised representation-guided image generation framework that enhances label-free medical image synthesis by leveraging self-supervised learning representations, enabling high-quality generation across different modalities without requiring labels or annotations. Our framework incorporates a Self-supervised Guidance Module that provides rich semantic knowledge during training and introduces a Guidance Representation Generator to bridge the train-inference disparity. Through extensive evaluation across four diverse medical datasets including brain MRI, chest X-ray, kidney CT, and eye glaucoma images, we demonstrate that RepMedGAN consistently achieves state-of-the-art results across multiple metrics and produces superior-quality medical images.

Ultrasound imaging serves as an important clinical diagnostic modality due to its non-invasive, radiation-free, and real-time capabilities. However, ultrasound images suffer from speckle noise that significantly compromises diagnostic accuracy and clinical interpretation. Traditional denoising methods are limited by speckle noise's signal-dependent nature, often removing important diagnostic features. While deep learning performs better, it requires large labelled datasets that are difficult to obtain due to privacy concerns and annotation costs. Self-supervised learning through masked image modeling (MIM) shows potential in addressing data scarcity, but conventional MIM, developed for high-level vision tasks, is unsuitable for low-level tasks like image denoising due to its framework architecture and learning strategy. To this end, we propose Image Denoising Masked Image Modeling (ID-MIM), the first MIM framework for ultrasound image denoising. ID-MIM incorporates a novel high-frequency oriented dual-branch masking and a specialized learning objective for noise reduction. Our encoder-only architecture features a multi-scale hierarchical transformer with dynamic skip connections, where the encoder directly performs denoising rather than relying on separate decoder reconstruction as in conventional MIM approaches. Extensive experiments demonstrate the superior performance of our ID-MIM framework across diverse noise scenarios, establishing new state-of-the-art results.

Atrial fibrillation (AF) is a significant risk factor for ischemic stroke recurrence, yet its diagnosis remains challenging through short-term heart monitoring due to its often paroxysmal and silent nature. Despite its diagnostic superiority, prolonged cardiac monitoring is typically impractical and not cost-effective for widespread implementation. We propose a novel AF risk stratification framework using a multimodal deep learning approach that integrates diffusion-weighted imaging (DWI) of the brain with clinical patient data. Our methodology combines convolutional neural networks (CNNs) for image analysis and gradient-boosted decision trees (GBDT) for clinical data, leveraging an innovative fusion strategy and an auxiliary loss function based on infarct location. The proposed approach achieves an area under the receiver operating characteristic (AUROC) of 89.18%, outperforming unimodal counterparts. This work contributes to the field by enabling AF risk stratification from brain DWI, utilizing weak supervision, and introducing a novel early and late-stage data fusion approach. Our method easily integrates with existing workflows and can identify high-risk individuals requiring intensive cardiac monitoring.

Self-supervised learning has achieved state-of-the-art performance in various tasks and applications. In computer vision, self-supervised learning often employs contrastive learning and masked image modeling, each with its limitations: contrastive learning heavily relies on strong data augmentation and large batch sizes, etc., while masked image modeling struggles to capture high-level semantics and discrimination. In this work, we introduce MAsked Contrastive Representation Learning (MACRL), a novel framework that integrates both paradigms through an asymmetric siamese network design. The online and momentum branches of the network receive asymmetric data augmentation operations and extract features through their encoders. The decoder in the online branch reconstructs the original image, while the projectors in both branches compute the contrastive loss. The online branch and the momentum branch are updated through gradient backpropagation and exponential moving average, respectively. MACRL jointly optimizes the reconstruction and the contrastive objectives to encourage representations with enhanced discrimination and semantics. Experimental results show that MACRL achieves competitive performance in downstream vision tasks, including image classification and semantic segmentation. Moreover, it demonstrates consistent performance across both large-scale and small-scale datasets.

Multiple-set canonical correlation analysis (mCCA) is a generalization of canonical correlation analysis (CCA) to three or more sets of variables. It aims to study the relationships between several sets of variables and it subsumes a number of interesting multivariate data analysis techniques as special cases. The quality and interpretability of the mCCA components are likely to be affected by the usefulness and relevance of each set of variables. Therefore, it is an important issue to identify each set of significant variables that are active in the relationships between sets. In this paper mCCA is extended to address the issue of variable set selection. Specifically a block sparse multiple set canonical correlation analysis (BSmCCA) algorithm is proposed to combine mCCA with ℓ2-norm type penalty in a unified framework. Within this framework sets of variables that are not necessarily relevant are removed. This makes BSmCCA a flexible method for analyzing for Multi-Subject functional magnetic resonance imaging (fMRI) data sets. The performances of the proposed BSmCCA algorithm are illustrated through on block design paradigm finger taping fMRI datasets.