Magnetic resonance imaging (MRI) provides diverse perspectives on anatomical structures, enabling multi-contrast super-resolution (SR) techniques that leverage complementary information across modalities to significantly enhance image quality. However, most existing multi-contrast SR methods are computationally intensive and lack lightweight solutions. In this study, we propose a novel lightweight progressive aggregation network (PAN) architecture for multi-contrast MRI SR. Our approach introduces multi-perception and residual feature aggregation mechanisms, which effectively capture and integrate anatomical details from low-resolution and reference images. Extensive experiments demonstrate that our lightweight method achieves superior efficiency, significantly outperforming other multi-contrast MRI SR methods in experiments, offering a promising solution for resource-constrained multi-contrast MRI super-resolution scenarios where computational efficiency is critical. The code can be found at https://github.com/Huaibei-normal-university-cv-laboratory/PAN .

ABSTRACT Depression, a prevalent mental disorder in modern society, significantly impacts people's daily lives. Recently, there have been advancements in developing automated diagnosis models for detecting depression. However, data scarcity, primarily due to privacy concerns, has posed a challenge. Traditional speech features have limitations in representing knowledge for depression diagnosis, and the complexity of deep learning algorithms necessitates substantial data support. Furthermore, existing multimodal methods based on neural networks overlook the heterogeneity gap between different modalities, potentially resulting in redundant information. To address these issues, we propose a multimodal depression detection model based on the Enhanced Cross‐Attention (ECA) Mechanism. This model effectively explores text‐speech interactions while considering modality heterogeneity. Data scarcity has been mitigated by fine‐tuning pre‐trained models. Additionally, we design a modal fusion module based on ECA, which emphasizes similarity responses and updates the weight of each modal feature based on the similarity information between modal features. Furthermore, for speech feature extraction, we have reduced the computational complexity of the model by integrating a multi‐window self‐attention mechanism with the Fourier transform. The proposed model is evaluated on the public dataset, DAIC‐WOZ, achieving an accuracy of 80.0% and an average F 1 value improvement of 4.3% compared with relevant methods.

ABSTRACT Image‐text retrieval faces challenges such as insufficient fusion of cross‐modal matching relationships, limited ability to distinguish between positive and negative samples, and inadequate feature representation. To address these issues, this paper proposes a new visual‐semantic interaction network (CFISRO). Building upon standard feature extractors, the network jointly models image and text features through a multi‐layer cross‐interaction module, capturing the rich semantic relationships between images and text. During the training process, a positive–negative sample similarity ranking optimization strategy is introduced to effectively reduce the distance between matching sample pairs while expanding the distance between non‐matching samples, enhancing the model's ability to distinguish. Furthermore, the network optimizes the image‐text matching classification performance by minimizing the difference between the predicted probability distribution and the true label distribution. Overall, CFISRO builds a collaborative optimization mechanism across multiple levels of image‐text representation learning, similarity modeling, and matching discrimination, significantly improving the representation quality and matching accuracy in cross‐modal retrieval tasks. Experimental results show that our method achieves excellent performance on benchmark datasets such as MS‐COCO and Flickr30K.