Recent advances in parameter-efficient transfer learning have demonstrated the utility of composing LoRA adapters from libraries of pretrained modules. However, most existing approaches rely on simple retrieval heuristics or uniform averaging, which overlook the latent structure of task relationships in representation space. We propose a new framework for adapter reuse that moves beyond retrieval, formulating adapter composition as a geometry-aware sparse reconstruction problem. Specifically, we represent each task by a latent prototype vector derived from the base model’s encoder and aim to approximate the target task prototype as a sparse linear combination of retrieved reference prototypes, under an L1-regularized optimization objective. The resulting combination weights are then used to blend the corresponding LoRA adapters, yielding a composite adapter tailored to the target task. This formulation not only preserves the local geometric structure of the task representation manifold, but also promotes interpretability and efficient reuse by selecting a minimal set of relevant adapters. We demonstrate the effectiveness of our approach across multiple domains—including medical image segmentation, medical report generation and image synthesis. Our results highlight the benefit of coupling retrieval with latent geometry-aware optimization for improved zero-shot generalization.

Echocardiography (ECHO) is essential for cardiac assessments, but its video quality and interpretation heavily relies on manual expertise, leading to inconsistent results from clinical and portable devices. ECHO video generation offers a solution by improving automated monitoring through synthetic data and generating high-quality videos from routine health data. However, existing models often face high computational costs, slow inference, and rely on complex conditional prompts that require experts' annotations. To address these challenges, we propose ECHOPulse, an ECG-conditioned ECHO video generation model. ECHOPulse introduces two key advancements: (1) it accelerates ECHO video generation by leveraging VQ-VAE tokenization and masked visual token modeling for fast decoding, and (2) it conditions on readily accessible ECG signals, which are highly coherent with ECHO videos, bypassing complex conditional prompts. To the best of our knowledge, this is the first work to use time-series prompts like ECG signals for ECHO video generation. ECHOPulse not only enables controllable synthetic ECHO data generation but also provides updated cardiac function information for disease monitoring and prediction beyond ECG alone. Evaluations on three public and private datasets demonstrate state-of-the-art performance in ECHO video generation across both qualitative and quantitative measures. Additionally, ECHOPulse can be easily generalized to other modality generation tasks, such as cardiac MRI, fMRI, and 3D CT generation. We will make the synthetic ECHO dataset, along with the code and model, publicly available upon acceptance.

Ensuring fairness in medical image segmentation is critical due to biases in imbalanced clinical data acquisition caused by demographic attributes (e.g., age, sex, race) and clinical factors (e.g., disease severity). To address these challenges, we introduce Distribution-aware Mixture of Experts (dMoE), inspired by optimal control theory. We provide a comprehensive analysis of its underlying mechanisms and clarify dMoE's role in adapting to heterogeneous distributions in medical image segmentation. Furthermore, we integrate dMoE into multiple network architectures, demonstrating its broad applicability across diverse medical image analysis tasks. By incorporating demographic and clinical factors, dMoE achieves state-of-the-art performance on two 2D benchmark datasets and a 3D in-house dataset. Our results highlight the effectiveness of dMoE in mitigating biases from imbalanced distributions, offering a promising approach to bridging control theory and medical image segmentation within fairness learning paradigms. The source code is available at https://github.com/tvseg/dMoE.

Despite achieving impressive results in general-purpose semantic segmentation with strong generalization on natural images, the Segment Anything Model (SAM) has shown less precision and stability in medical image segmentation. In particular, the original SAM architecture is designed for 2D natural images and is therefore not support to handle three-dimensional information, which is particularly important for medical imaging modalities that are often volumetric or video data. In this paper, we introduce MediViSTA, a parameter-efficient fine-tuning method designed to adapt the vision foundation model for medical video, with a specific focus on echocardiography segmentation. To achieve spatial adaptation, we propose a frequency feature fusion technique that injects spatial frequency information from a CNN branch. For temporal adaptation, we integrate temporal adapters within the transformer blocks of the image encoder. Using a fine-tuning strategy, only a small subset of pre-trained parameters is updated, allowing efficient adaptation to echocardiography data. The effectiveness of our method has been comprehensively evaluated on three datasets, comprising two public datasets and one multi-center in-house dataset. Our method consistently outperforms various state-of-the-art approaches without using any prompts. Furthermore, our model exhibits strong generalization capabilities on unseen datasets, surpassing the second-best approach by 2.15% in Dice and 0.09 in temporal consistency. The results demonstrate the potential of MediViSTA to significantly advance echocardiography video segmentation, offering improved accuracy and robustness in cardiac assessment applications.

Echocardiography is the first-line non-invasive cardiac imaging modality, providing rich spatio-temporal information on cardiac anatomy and physiology. Recently, foundation model trained on extensive and diverse datasets has shown strong performance in various downstream tasks. However, translating foundation models into the medical imaging domain remains challenging due to domain differences between medical and natural images, the lack of diverse patient and disease datasets. In this paper, we introduce EchoFM, a general-purpose vision foundation model for echocardiography trained on a large-scale dataset of over 20 million echocardiographic images from 6,500 patients. To enable effective learning of rich spatio-temporal representations from periodic videos, we propose a novel self-supervised learning framework based on a masked autoencoder with a spatio-temporal consistent masking strategy and periodic-driven contrastive learning. The learned cardiac representations can be readily adapted and fine-tuned for a wide range of downstream tasks, serving as a strong and flexible backbone model. We validate EchoFM through experiments across key downstream tasks in the clinical echocardiography workflow, leveraging public and multi-center internal datasets. EchoFM consistently outperforms SOTA methods, demonstrating superior generalization capabilities and flexibility. The code and checkpoints are available at: https://github.com/SekeunKim/EchoFM.git.

Prediction of abdominal aortic aneurysm (AAA) growth is of essential importance for the early treatment and surgical intervention of AAA. Capturing key features of vascular growth, such as blood flow and intraluminal thrombus (ILT) accumulation play a crucial role in uncovering the intricated mechanism of vascular adaptation, which can ultimately enhance AAA growth prediction capabilities. However, local correlations between hemodynamic metrics, biological and morphological characteristics, and AAA growth rates present high inter-patient variability that results in that the temporal-spatial biochemical and mechanical processes are still not fully understood. Hence, this study aims to integrate the physics-based knowledge with deep learning with a patch-based convolutional neural network (CNN) approach by incorporating important multiphysical features relating to its pathogenesis for validating its impact on AAA growth prediction. For this task, we observe that the unstructured multiphysical features cannot be directly employed in the kernel-based CNN. To tackle this issue, we propose a parameterization of features to leverage the spatio-temporal relations between multiphysical features. The proposed architecture was tested on different combinations of four features including radius, intraluminal thrombus thickness, time-average wall shear stress, and growth rate from 54 patients with 5-fold cross-validation with two metrics, a root mean squared error (RMSE) and relative error (RE). We conduct extensive experiments on AAA patients, the results show the effect of leveraging multiphysical features and demonstrate the superiority of the presented architecture to previous state-of-the-art methods in AAA growth prediction.