Ultrasound (US) imaging plays a crucial role in diagnosing heart and pelvic diseases, where sonographers tend to evaluate dynamic motion and structure. However, the scarcity of US videos for rare cases limitstraining opportunities for novice sonographers and deep learning models, hindering detection rates and clinical diagnostic applications. US video synthesis is a promising solution to this issue. Nevertheless, accurately imitating the intricate motion of the anatomy while preserving image fidelity presents asignificant challenge. In this work, we propose OnUVS, a novel online feature-decoupling framework for high-fidelity US video synthesis. First, to simulate realistic motion, we incorporate keypoints into anatomical learning through a weakly supervised training approach, which enhances motion representation and minimizes the need for fully annotated data. Second, we implement a dual-decoder generator that effectively balances content and textural features of generated frames, significantly enhancing the image fidelity of US videos. Third, a multi-scale discriminator further refines the sharpness and fine details, ensuring high-fidelity video synthesis. Fourth, an online learning strategy is designed to smooth coherence between frames by constraining the keypoint trajectories during inference. Validation on echocardiographic and pelvic floor US datasets demonstrates that OnUVS outperforms existing methods, achieving a 22.08% improvement in motion consistency (FVD) and 25.04% in image fidelity (FID).

Brain lesion segmentation is crucial for neurological disease research and diagnosis. As different types of lesions exhibit distinct characteristics on different imaging modalities, segmentation methods are typically developed in a task-specific manner, where each segmentation model is tailored to a specific lesion type and modality. However, the use of task-specific models requires predetermination of the lesion type and imaging modality, which complicates their deployment in real-world scenarios. In this work, we propose a universal foundation model for brain lesion segmentation on magnetic resonance imaging (MRI), which can automatically segment different types of brain lesions given input of various MRI modalities. We develop a novel Mixture of Modality Experts (MoME) framework with multiple expert networks attending to different imaging modalities. A hierarchical gating network is proposed to combine the expert predictions and foster expertise collaboration. Moreover, to avoid the degeneration of each expert network, we introduce a curriculum learning strategy during training to preserve the specialisation of each expert. In addition to MoME, to handle the combination of multiple input modalities, we propose MoME+, which uses a soft dispatch network for input modality routing. We evaluated the proposed method on nine brain lesion datasets, encompassing five imaging modalities and eight lesion types. The results show that our model outperforms state-of-the-art universal models for brain lesion segmentation and achieves promising generalisation performance onto unseen datasets.

Two key questions in cardiac image analysis are to assess the anatomy and motion of the heart from images; and to understand how they are associated with non-imaging clinical factors such as gender, age and diseases. While the first question can often be addressed by image segmentation and motion tracking algorithms, our capability to model and answer the second question is still limited. In this work, we propose a novel conditional generative model to describe the 4D spatio-temporal anatomy of the heart and its interaction with non-imaging clinical factors. The clinical factors are integrated as the conditions of the generative modelling, which allows us to investigate how these factors influence the cardiac anatomy. We evaluate the model performance in mainly two tasks, anatomical sequence completion and sequence generation. The model achieves high performance in anatomical sequence completion, comparable to or outperforming other state-of-the-art generative models. In terms of sequence generation, given clinical conditions, the model can generate realistic synthetic 4D sequential anatomies that share similar distributions with the real data. The code and the trained generative model are available at https://github.com/MengyunQ/CHeart.

Multimodal analysis of placental ultrasound (US) and microflow imaging (MFI) could greatly aid in the early diagnosis and interventional treatment of placental insufficiency (PI), ensuring a normal pregnancy. Existing multimodal analysis methods have weaknesses in multimodal feature representation and modal knowledge definitions and fail on incomplete datasets with unpaired multimodal samples. To address these challenges and efficiently leverage the incomplete multimodal dataset for accurate PI diagnosis, we propose a novel graph-based manifold regularization learning (MRL) framework named GMRLNet. It takes US and MFI images as input and exploits their modality-shared and modality-specific information for optimal multimodal feature representation. Specifically, a graph convolutional-based shared and specific transfer network (GSSTN) is designed to explore intra-modal feature associations, thus decoupling each modal input into interpretable shared and specific spaces. For unimodal knowledge definitions, graph-based manifold knowledge is introduced to describe the sample-level feature representation, local inter-sample relations, and global data distribution of each modality. Then, an MRL paradigm is designed for inter-modal manifold knowledge transfer to obtain effective cross-modal feature representations. Furthermore, MRL transfers the knowledge between both paired and unpaired data for robust learning on incomplete datasets. Experiments were conducted on two clinical datasets to validate the PI classification performance and generalization of GMRLNet. State-of-the-art comparisons show the higher accuracy of GMRLNet on incomplete datasets. Our method achieves 0.913 AUC and 0.904 balanced accuracy (bACC) for paired US and MFI images, as well as 0.906 AUC and 0.888 bACC for unimodal US images, illustrating its application potential in PI CAD systems.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and ultrasound (US), which are two common modalities for clinical breast tumor diagnosis besides Mammograms, can provide different and complementary information for the same tumor regions. Although many machine learning methods have been proposed for breast tumor classification based on either single modality, it remains unclear how to further boost the classification performance by utilizing paired multi-modality information with different dimensions. In this paper, we propose MRI-US multi-modality network (MUM-Net) to classify breast tumor into different subtypes based on 3D MR and 2D US images. The key insight of MUM-Net is that we explicitly distill modality-agnostic features for tumor classification. Specifically, we first adopt a discrimination-adaption module to decompose features into modality-agnostic and modality-specific ones with min-max training strategies. Then, we propose a feature fusion module to increase the compactness of the modality-agnostic features by utilizing an affinity matrix with nearest neighbour selection. We build a paired MRI-US breast tumor classification dataset containing 502 cases with three clinical indicators to validate the proposed method. In three tasks including lymph node metastasis, histological grade and Ki-67 level, MUM-Net achieves AUC scores of 0.8581, 0.8965 and 0.8577, outperforming other counterparts which are based on single task or single modality by a wide margin. In addition, we find that the extracted modality-agnostic features can help the network focus on the tumor regions in both modalities.