Author name cluster

Chun Li

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.

8 papers

2 author rows

EAAI Journal 2025 Journal Article

A diffusion model based on multi-scale spatial Mamba for medical image segmentation

Chun Li
Qiule Sun
Muqing Zhang
Jianxin Zhang

Medical image segmentation plays a key role in disease diagnosis, treatment planning, and monitoring disease progression. Recently, denoising diffusion models have shown significant promise in generating accurate pixel-level semantic representations. In this study, we introduce a diffusion model based on multi-scale spatial mamba (MSM-Diff) designed for precise medical image segmentation. MSM-Diff integrates the strengths of diffusion models, Mamba architecture, and convolutional neural networks to efficiently capture both global and local contextual information from complex volumetric data. The core of MSM-Diff is the Mamba-based U-shaped feature encoder (MUFE), which combines the three-dimensional multi-scale spatial Mamba model (MS-Mamba) with extracted features for enhanced multi-scale and global feature extraction. By using the mamba architecture, the model maintains linear computational complexity. Additionally, MSM-Diff incorporates a multi-scale gated spatial convolution (MS-GSC) module within MUFE to further refine spatial feature representations. Extensive evaluations of three public datasets demonstrate that MSM-Diff consistently outperforms current methods, achieving state-of-the-art performance in DSC and HD95. This model provides a robust solution for medical image segmentation by effectively capturing global context and accurately delineating boundaries, thereby improving diagnostic and treatment planning outcomes for radiologists and clinicians.

Details DOI

NeurIPS Conference 2025 Conference Paper

Federated Dialogue-Semantic Diffusion for Emotion Recognition under Incomplete Modalities

Xihang Qiu
Jiarong Cheng
Yuhao Fang
WanPeng Zhang
Yao Lu
Ye Zhang
Chun Li

Multimodal Emotion Recognition in Conversations (MERC) enhances emotional understanding through the fusion of multimodal signals. However, unpredictable modality absence in real-world scenarios significantly degrades the performance of existing methods. Conventional missing-modality recovery approaches, which depend on training with complete multimodal data, often suffer from semantic distortion under extreme data distributions, such as fixed-modality absence. To address this, we propose the Federated Dialogue-guided and Semantic-Consistent Diffusion (FedDISC) framework, pioneering the integration of federated learning into missing-modality recovery. By federated aggregation of modality-specific diffusion models trained on clients and broadcasting them to clients missing corresponding modalities, FedDISC overcomes single-client reliance on modality completeness. Additionally, the DISC-Diffusion module ensures consistency in context, speaker identity, and semantics between recovered and available modalities, using a Dialogue Graph Network to capture conversational dependencies and a Semantic Conditioning Network to enforce semantic alignment. We further introduce a novel Alternating Frozen Aggregation strategy, which cyclically freezes recovery and classifier modules to facilitate collaborative optimization. Extensive experiments on the IEMOCAP, CMUMOSI, and CMUMOSEI datasets demonstrate that FedDISC achieves superior emotion classification performance across diverse missing modality patterns, outperforming existing approaches.

PDF Details

JBHI Journal 2025 Journal Article

FedKDC: Consensus-Driven Knowledge Distillation for Personalized Federated Learning in EEG-Based Emotion Recognition

Xihang Qiu
Wanyong Qiu
Ye Zhang
Kun Qian
Chun Li
Bin Hu
Björn W. Schuller
Yoshiharu Yamamoto

Federated learning (FL) has gained prominence in electroencephalogram (EEG)-based emotion recognition because of its ability to enable secure collaborative training without centralized data. However, traditional FL faces challenges due to model and data heterogeneity in smart healthcare settings. For example, medical institutions have varying computational resources, which creates a need for personalized local models. Moreover, EEG data from medical institutions typically face data heterogeneity issues stemming from limitations in participant availability, ethical constraints, and cultural differences among subjects, which can slow model convergence and degrade model performance. To address these challenges, we propose FedKDC, a novel FL framework that incorporates clustered knowledge distillation (CKD). This method introduces a consensus-based distributed learning mechanism to facilitate the clustering process. It then enhances the convergence speed through intraclass distillation and reduces the negative impact of heterogeneity through interclass distillation. Additionally, we introduce a DriftGuard mechanism to mitigate client drift, along with an entropy reducer to decrease the entropy of aggregated knowledge. The framework is validated on the SEED, SEED-IV, SEED-FRA, and SEED-GER datasets, demonstrating its effectiveness in scenarios where both the data and the models are heterogeneous. Experimental results show that FedKDC outperforms other FL frameworks in emotion recognition, achieving a maximum average accuracy of 85. 2%, and in convergence efficiency, with faster and more stable convergence.

Details DOI

EAAI Journal 2025 Journal Article

Length-aware center loss for sequence to sequence Thai scene text recognition

Hongjian Zhan
Chun Li
Bing Yin
Yue Lu

Thai scene text recognition is a challenging task because Thai can be written in both horizontal and vertical directions, allowing characters to be stacked vertically. To address this issue, our previous work combined vertically stacked characters to create new characters. However, this strategy introduced many similar characters. In this paper, we further investigate this problem and propose the Length-aware Center Loss (LC) for Thai scene text recognition. The original center loss was designed for single object recognition tasks. When applied to multi-label tasks like text recognition, center loss is only effective when the lengths of the labels and prediction results are consistent. This can lead to an extreme case where all images receive incorrect predicted text lengths to minimize loss, severely interfering with the recognition process. Therefore, we propose the Length-aware Center Loss for text recognition. We also design the Length Supervision Module (LSM) and the Feature Clustering Module (FCM) to work alongside the LC loss. LSM predicts text length to provide additional supervision signals, while FCM aims to improve recognition performance by minimizing the distance between the features of corresponding class centers. Since there is no publicly available Thai scene text dataset, we have collected a new dataset containing more than 170, 000 samples. Extensive experiments conducted on this dataset show that our method achieves superior performance in both string-level and character-level accuracy compared to other methods.

Details DOI

NeurIPS Conference 2025 Conference Paper

Proper Hölder-Kullback Dirichlet Diffusion: A Framework for High Dimensional Generative Modeling

WanPeng Zhang
Yuhao Fang
Xihang Qiu
Jiarong Cheng
Jialong Hong
Bin Zhai
Qing Zhou
Yao Lu

Diffusion-based generative models have long depended on Gaussian priors, with little exploration of alternative distributions. We introduce a Proper Hölder-Kullback Dirichlet framework that uses time-varying multiplicative transformations to define both forward and reverse diffusion processes. Moving beyond conventional reweighted evidence lower bounds (ELBO) or Kullback–Leibler upper bounds (KLUB), we propose two novel divergence measures: the Proper Hölder Divergence (PHD) and the Proper Hölder–Kullback (PHK) divergence, the latter designed to restore symmetry missing in existing formulations. When optimizing our Dirichlet diffusion model with PHK, we achieve a Fréchet Inception Distance (FID) of 2. 78 on unconditional CIFAR-10. Comprehensive experiments on natural-image datasets validate the generative strengths of model and confirm PHK’s effectiveness in model training. These contributions expand the diffusion-model family with principled non-Gaussian processes and effective optimization tools, offering new avenues for versatile, high-fidelity generative modeling.

PDF Details

JBHI Journal 2024 Journal Article

Semi-Supervised Disease Classification Based on Limited Medical Image Data

Yan Zhang
Chun Li
Zhaoxia Liu
Ming Li

Inrecent years, significant progress has been made in the field of learning from positive and unlabeled examples (PU learning), particularly in the context of advancing image and text classification tasks. However, applying PU learning to semi-supervised disease classification remains a formidable challenge, primarily due to the limited availability of labeled medical images. In the realm of medical image-aided diagnosis algorithms, numerous theoretical and practical obstacles persist. The research on PU learning for medical image-assisted diagnosis holds substantial importance, as it aims to reduce the time spent by professional experts in classifying images. Unlike natural images, medical images are typically accompanied by a scarcity of annotated data, while an abundance of unlabeled cases exists. Addressing these challenges, this paper introduces a novel generative model inspired by Hölder divergence, specifically designed for semi-supervised disease classification using positive and unlabeled medical image data. In this paper, we present a comprehensive formulation of the problem and establish its theoretical feasibility through rigorous mathematical analysis. To evaluate the effectiveness of our proposed approach, we conduct extensive experiments on five benchmark datasets commonly used in PU medical learning: BreastMNIST, PneumoniaMNIST, BloodMNIST, OCTMNIST, and AMD. The experimental results clearly demonstrate the superiority of our method over existing approaches based on KL divergence. Notably, our approach achieves state-of-the-art performance on all five disease classification benchmarks. By addressing the limitations imposed by limited labeled data and harnessing the untapped potential of unlabeled medical images, our novel generative model presents a promising direction for enhancing semi-supervised disease classification in the field of medical image analysis.

Details DOI

YNIMG Journal 2021 Journal Article

Inter-individual variability in structural brain development from late childhood to young adulthood

Kathryn L. Mills
Kimberly D. Siegmund
Christian K. Tamnes
Lia Ferschmann
Lara M. Wierenga
Marieke G.N. Bos
Beatriz Luna
Chun Li

A fundamental task in neuroscience is to characterize the brain's developmental course. While replicable group-level models of structural brain development from childhood to adulthood have recently been identified, we have yet to quantify and understand individual differences in structural brain development. The present study examined inter-individual variability and sex differences in changes in brain structure, as assessed by anatomical MRI, across ages 8.0-26.0 years in 269 participants (149 females) with three time points of data (807 scans), drawn from three longitudinal datasets collected in the Netherlands, Norway, and USA. We further investigated the relationship between overall brain size and developmental changes, as well as how females and males differed in change variability across development. There was considerable inter-individual variability in the magnitude of changes observed for all examined brain measures. The majority of individuals demonstrated decreases in total gray matter volume, cortex volume, mean cortical thickness, and white matter surface area in mid-adolescence, with more variability present during the transition into adolescence and the transition into early adulthood. While most individuals demonstrated increases in white matter volume in early adolescence, this shifted to a majority demonstrating stability starting in mid-to-late adolescence. We observed sex differences in these patterns, and also an association between the size of an individual's brain structure and the overall rate of change for the structure. The present study provides new insight as to the amount of individual variance in changes in structural morphometrics from late childhood to early adulthood in order to obtain a more nuanced picture of brain development. The observed individual- and sex-differences in brain changes also highlight the importance of further studying individual variation in developmental patterns in healthy, at-risk, and clinical populations.

Details DOI

ICRA Conference 1999 Conference Paper

A Decentralized Approach to the Conflict-Free Motion Planning for Multiple Mobile Robots

Chun Li
Zhiqiang Zheng
Wensen Chang

Presents a decentralized approach to the conflict-free motion planning for multiple mobile robots. We decompose the problem into global path planning and local path planning. AI techniques are used to solve the global planning problem and we obtain an optimal global path. From the point of view of combining AI techniques with a real-time control technique; a dynamic sub-goal algorithm is put forward to solve the local planning problem for the multi-robot system. The simulation results show the efficiency of the algorithm for real-time multi-robot cooperation.

Details