• IMH-Net boosts segmentation accuracy via salient-first modeling and hypergraph fusion • IA-Mamba prioritizes salient regions and preserves fine-grained local details • CSCEM boosts channel spatial complementarity and suppresses crossmodal noise • CHB capture high-order skip connection dependencies to recover details in decoding • Experiments show IMH-Net beats prior methods and stays SOTAcompetitive Precise multimodal tumor segmentation is essential for radiotherapy target contouring, surgical planning, and therapeutic efficacy evaluation. PET provides metabolic activity information, whereas CT offers detailed anatomical structures; their complementarity improves segmentation reliability in complex cases. However, existing sequence-modeling schemes are susceptible to order bias induced by a fixed scanning order, and cross-modal fusion and skip-connection interactions often remain at low-order, coarse-grained levels, making it difficult to jointly achieve salient-region–prioritized modeling, noise suppression, and high-order semantic coupling. To address this, we propose IMH-Net, an automatic multimodal tumor segmentation network based on importance-aware Mamba and hypergraph modeling. The proposed network includes three core components: (1) importance-aware Mamba (IA-Mamba), which estimates patch importance in the encoder stage and dynamically reshuffles the scan order to model salient regions first. (2) The Cross-modal Spatial Channel Enhancement Module (CSCEM) performs cross-modal collaborative enhancement in both channel and spatial dimensions at the bottleneck, emphasizing complementary semantics while suppressing redundant conflicts. (3) The Cross-modal Hypergraph Bridge (CHB) constructs intra- and inter-modality hyperedges at skip connections and leverages hypergraph convolution and hypergraph attention to enable stable high-order interactions and feature coupling. Comprehensive experiments on the public STS, Hecktor 2022, and ECPC datasets validate both the effectiveness of the proposed modules and their complementary synergy. IMH-Net achieves Dice scores of 81.82%, 80.86%, and 91.40% on STS, Hecktor 2022, and ECPC datasets, respectively, outperforming state-of-the-art (SOTA) multimodal segmentation methods in overall performance.

Deep learning models often struggle with class imbalance and low-resolution medical images, where critical spatial details and minority-class features are underrepresented. We introduce the Adaptive Distribution-aware Vision Transformer (AdaptiveViT), a novel hybrid CNN-Transformer architecture that unifies fine-grained local feature extraction with global contextual modelling. AdaptiveViT incorporates a distribution-aware modulation mechanism that adaptively adjusts feature emphasis according to the severity of class imbalance. In addition, a Distribution-aware Adaptive (DA) Loss incorporates the dataset imbalance ratio into an adaptive focusing scheme, enhancing minority-class sensitivity. Experiments on five skin lesion datasets with varying image resolutions and imbalance ratios (as high as 1:10 for melanoma versus non-melanoma) demonstrate that AdaptiveViT consistently outperforms state-of-the-art CNN, Transformer, and hybrid baselines in F1 and AUC, while maintaining stable convergence across imbalance levels. Validation on gastrointestinal endoscopy datasets further demonstrates AdaptiveViT's domain-agnostic generalisation beyond skin lesion data, which share similar imbalance characteristics. All experiments are conducted using patient-disjoint splits, with a threshold-free evaluation protocol to ensure fair, unbiased, and clinically reliable comparisons. Overall, AdaptiveViT establishes a hybrid framework for medical image classification under class imbalance and image-resolution variability. The code is available at https://github.com/mmu-dermatology-research/AdaptiveViT.

Deep learning-based image registration methods have shown state-of-the-art performance and rapid inference speeds. Despite these advances, many existing approaches fall short in capturing spatially varying information in non-local regions of feature maps due to the reliance on spatially-shared convolution kernels. This limitation leads to suboptimal estimation of deformation fields. In this paper, we propose a 3D Spatial-Awareness Convolution Block (SACB) to enhance the spatial information within feature representations. Our SACB estimates the spatial clusters within feature maps by leveraging feature similarity and subsequently parameterizes the adaptive convolution kernels across diverse regions. This adaptive mechanism generates the convolution kernels (weights and biases) tailored to spatial variations, thereby enabling the network to effectively capture spatially varying information. Building on SACB, we introduce a pyramid flow estimator (named SACB-Net) that integrates SACBs to facilitate multi-scale flow composition, particularly addressing large deformations. Experimental results on the brain IXI and LPBA datasets as well as Abdomen CT datasets demonstrate the effectiveness of SACB and the superiority of SACB-Net over the state-of-the-art learning-based registration methods. The code is available at https://github.com/x-xc/SACB_Net.

In unsupervised medical image registration, encoder-decoder architectures are widely used to predict dense, full-resolution displacement fields from paired images. Despite their popularity, we question the necessity of making both the encoder and decoder learnable. To address this, we propose LessNet, a simplified network architecture with only a learnable decoder, while completely omitting a learnable encoder. Instead, LessNet replaces the encoder with simple, handcrafted features, eliminating the need to optimize encoder parameters. This results in a compact, efficient, and decoder-only architecture for 3D medical image registration. We evaluate our decoder-only LessNet on five registration tasks: 1) inter-subject brain registration using the OASIS-1 dataset, 2) atlas-based brain registration using the IXI dataset, 3) cardiac ES-ED registration using the ACDC dataset, 4) inter-subject abdominal MR registration using the CHAOS dataset, and 5) multi-study, multi-site brain registration using images from 13 public datasets. Our results demonstrate that LessNet can effectively and efficiently learn both dense displacement and diffeomorphic deformation fields. Furthermore, our decoder-only LessNet can achieve comparable registration performance to benchmarking methods such as VoxelMorph and TransMorph, while requiring significantly fewer computational resources. Our code and pre-trained models are available at https://github.com/xi-jia/LessNet.

The Alternating Direction Method of Multipliers (ADMM) has gained significant attention across a broad spectrum of machine learning applications. Incorporating the over-relaxation technique shows potential for enhancing the convergence rate of ADMM. However, determining optimal algorithmic parameters, including both the associated penalty and relaxation parameters, often relies on empirical approaches tailored to specific problem domains and contextual scenarios. Incorrect parameter selection can significantly hinder ADMM's convergence rate. To address this challenge, in this paper we first propose a general approach to optimize the value of penalty parameter, followed by a novel closed-form formula to compute the optimal relaxation parameter in the context of linear quadratic problems (LQPs). We then experimentally validate our parameter selection methods through random instantiations and diverse imaging applications, encompassing diffeomorphic image registration, image deblurring, and MRI reconstruction.

Unsupervised image registration commonly adopts U-Net style networks to predict dense displacement fields in the full-resolution spatial domain. For high-resolution volumetric image data, this process is however resource-intensive and time-consuming. To tackle this problem, we propose the Fourier-Net, replacing the expansive path in a U-Net style network with a parameter-free model-driven decoder. Specifically, instead of our Fourier-Net learning to output a full-resolution displacement field in the spatial domain, we learn its low-dimensional representation in a band-limited Fourier domain. This representation is then decoded by our devised model-driven decoder (consisting of a zero padding layer and an inverse discrete Fourier transform layer) to the dense, full-resolution displacement field in the spatial domain. These changes allow our unsupervised Fourier-Net to contain fewer parameters and computational operations, resulting in faster inference speeds. Fourier-Net is then evaluated on two public 3D brain datasets against various state-of-the-art approaches. For example, when compared to a recent transformer-based method, named TransMorph, our Fourier-Net, which only uses 2.2% of its parameters and 6.66% of the multiply-add operations, achieves a 0.5% higher Dice score and an 11.48 times faster inference speed. Code is available at https://github.com/xi-jia/Fourier-Net.

In this work, we investigate image registration in a variational framework and focus on regularization generality and solver efficiency. We first propose a variational model combining the state-of-the-art sum of absolute differences (SAD) and a new arbitrary order total variation regularization term. The main advantage is that this variational model preserves discontinuities in the resultant deformation while being robust to outlier noise. It is however non-trivial to optimize the model due to its non-convexity, non-differentiabilities, and generality in the derivative order. To tackle these, we propose to first apply linearization to the problem to formulate a convex objective function and then break down the resultant convex optimization into several point-wise, closed-form subproblems using a fast, over-relaxed alternating direction method of multipliers (ADMM). With our proposed algorithm, we show that solving higher-order variational formulations is similar to solving their lower-order counterparts. Extensive experiments show that our ADMM is significantly more efficient than both the subgradient and primal-dual algorithms particularly when higher-order derivatives are used, and that our new models outperform state-of-the-art methods based on deep learning and free-form deformation. Our code implemented in both Matlab and Pytorch is publicly available at https://github.com/j-duan/AOTV.

Human epidermal growth factor receptor 2 (HER2) is an important prognostic and predictive factor which is overexpressed in 15–20% of breast cancer (BCa). The determination of its status is a key clinical decision making step for selection of treatment regimen and prognostication. HER2 status is evaluated using transcriptomics or immunohistochemistry (IHC) through in-situ hybridisation (ISH) which incurs additional costs and tissue burden and is prone to analytical variabilities in terms of manual observational biases in scoring. In this study, we propose a novel graph neural network (GNN) based model (SlideGraph+) to predict HER2 status directly from whole-slide images of routine Haematoxylin and Eosin (H&E) stained slides. The network was trained and tested on slides from The Cancer Genome Atlas (TCGA) in addition to two independent test datasets. We demonstrate that the proposed model outperforms the state-of-the-art methods with area under the ROC curve (AUC) values > 0.75 on TCGA and 0.80 on independent test sets. Our experiments show that the proposed approach can be utilised for case triaging as well as pre-ordering diagnostic tests in a diagnostic setting. It can also be used for other weakly supervised prediction problems in computational pathology. The SlideGraph+ code repository is available at https://github.com/wenqi006/SlideGraph along with an IPython notebook showing an end-to-end use case at https://github.com/TissueImageAnalytics/tiatoolbox/blob/develop/examples/full-pipelines/slide-graph.ipynb.