Author name cluster

Dengdi Sun

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

2 author rows

JBHI Journal 2026 Journal Article

Attention-based Multimodal Spatiotemporal Enhanced Interaction Network For Major Depressive Disorder Detection

Changxu Dong
Xinwei Liu
Shuoqiu Gan
Zongyun Gu
Bin Luo
DIRECT Consortium
Dengdi Sun

Although deep learning models have shown promising results in detecting major depressive disorder (MDD), two main limitations remain: insufficient exploitation of interactive information across multimodal brain networks and a lack of adaptive mechanisms for capturing crucial spatiotemporal dependencies among brain regions. To address these challenges, we propose the Attention-based Multimodal Spatiotemporal Enhanced Interaction Network (AM-SEIN) for MDD detection. Specifically, to tackle the first challenge, we integrate structural information from 3D structural magnetic resonance imaging (sMRI) with functional temporal data from functional magnetic resonance imaging (fMRI). Additionally, we design the Cross-Modal Interaction Network (CMIN) and fusion layer to enhance mutual information aggregation and facilitate interactive guidance between the two modalities. For the second challenge, we develop an attention-based adaptive spatiotemporal feature-extracting architecture for both modalities, incorporating the fMRI-based Adaptive Spatiotemporal Fusion (fASF) and the sMRI-based Regional-Level Content-Dependent (sRLCD) modules. This approach enables the effective encoding of inter-regional interactions relevant to MDD detection. Finally, the proposed AM-SEIN is evaluated on the Rest-meta-MDD(RMM) and Rest-meta-MDD-V2(RMM-V2) datasets, achieving state-of-the-art performance.

Details DOI

NeurIPS Conference 2025 Conference Paper

Segment Anything Model Meets Semi-supervised Medical Image Segmentation: A Novel Perspective

Haifeng Zhao
HaiYang Li
Lei-Lei Ma
Dengdi Sun

The scarcity of annotated medical imaging data has driven significant progress in semi-supervised learning to alleviate reliance on expensive expert labeling. While foundational vision models such as the Segment Anything Model (SAM) exhibit robust generalization in generic segmentation tasks, their direct application to medical images often results in suboptimal performance. To address this challenge, in this work, we propose a novel fully SAM-based semi-supervised medical image segmentation framework and develop the corresponding knowledge distillation-based learning strategy. Specifically, we first employ an efficient SAM variant as the backbone network of the semi‑supervised framework and update the default prompt embedding of SAM to unleash its full potential. Then, we utilize an original SAM, which is rich in prior knowledge, as the teacher to optimize our efficient student SAM backbone through hierarchical knowledge distillation and a dynamic loss weighting strategy. Extensive experiments on various medical datasets demonstrate that our method outperforms state-of-the-art semi-supervised segmentation approaches. Especially, our model requires less than 10% of the parameter size of the original SAM, enabling substantially lower deployment and storage overhead in real-world clinical settings.

PDF Details

EAAI Journal 2025 Journal Article

Semantic knowledge transfer for semi-supervised medical image segmentation

Shiwei Zhou
Haifeng Zhao
Leilei Ma
Dengdi Sun

In semi-supervised medical image segmentation, due to possible differences in information content and distribution between labeled and unlabeled data, dealing with the two separately usually prevents knowledge transfer from labeled to unlabeled data. This prevents the model from effectively sharing learned information between the two types of data. To alleviate this problem, we train labeled and unlabeled data as a whole. Semantic mixing of labeled and unlabeled data is achieved by selecting and exchanging some of the region images of both through a mask to generate complementary input views. In addition, due to the limited labeled data, the unlabeled data has a weak ability to distinguish categories in the feature space. Traditional methods rely on pixel positions to generate positive and negative samples for contrastive learning to solve this problem, but relying on pixel position sampling can easily lead to semantic inconsistency, which affects the effect of feature learning; therefore, to address this problem, we propose an innovative labeled data-guided inter-class contrastive learning strategy, which extracts the category features from labeled and unlabeled data and exploits the accurate category information in the labeled data to guide contrastive learning, while introducing a similarity-based ranking weighting mechanism. Combining the two designs, we propose a new semantic knowledge transfer framework for semi-supervised medical image segmentation. Experiments demonstrate a significant improvement in our model compared to State of the Art (SOTA) on the Automatic Cardiac Diagnosis Challenge (ACDC) dataset and the Left Atrium (LA) dataset.

Details DOI

ECAI Conference 2023 Conference Paper

Semantic-Aware Dual Contrastive Learning for Multi-Label Image Classification

Leilei Ma 0002
Dengdi Sun
Lei Wang 0095
Haifeng Zhao 0001
Bin Luo 0001

Extracting image semantics effectively and assigning corresponding labels to multiple objects or attributes for natural images is challenging due to the complex scene contents and confusing label dependencies. Recent works have focused on modeling label relationships with graph and understanding object regions using class activation maps (CAM). However, these methods ignore the complex intra- and inter-category relationships among specific semantic features, and CAM is prone to generate noisy information. To this end, we propose a novel semantic-aware dual contrastive learning framework that incorporates sample-to-sample contrastive learning (SSCL) as well as prototype-to-sample contrastive learning (PSCL). Specifically, we leverage semantic-aware representation learning to extract category-related local discriminative features and construct category prototypes. Then based on SSCL, label-level visual representations of the same category are aggregated together, and features belonging to distinct categories are separated. Meanwhile, we construct a novel PSCL module to narrow the distance between positive samples and category prototypes and push negative samples away from the corresponding category prototypes. Finally, the discriminative label-level features related to the image content are accurately captured by the joint training of the above three parts. Experiments on five challenging large-scale public datasets demonstrate that our proposed method is effective and outperforms the state-of-the-art methods. Code and supplementary materials are released on https: //github. com/yu-gi-oh-leilei/SADCL.

Details

IJCAI Conference 2011 Conference Paper

Angular Decomposition

Dengdi Sun
Chris Ding
Bin Luo
Jin Tang

Dimensionality reduction plays a vital role in pattern recognition. However, for normalized vector data, existing methods do not utilize the fact that the data is normalized. In this paper, we propose to employ an Angular Decomposition of the normalized vector data which corresponds to embedding them on a unit surface. On graph data for similarity/kernel matrices with constant diagonal elements, we propose the Angular Decomposition of the similarity matrices which corresponds to embedding objects on a unit sphere. In these angular embeddings, the Euclidean distance is equivalent to the cosine similarity. Thus data structures best described in the cosine similarity and data structures best captured by the Euclidean distance can both be effectively detected in our angular embedding. We provide the theoretical analysis, derive the computational algorithm, and evaluate the angular embedding on several datasets. Experiments on data clustering demonstrate that our method can provide a more discriminative subspace.

PDF Details DOI