ZDZhengxu Duan

Zhengxu Duan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Zhengxu Duan has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • DM-PAFNet: A Dual-Stage Deep Learning Framework for Parkinson's Disease Staging Prediction via Multi-modal MRI and Clinical Prior Fusion

    by SenYuan Li, Mei Yang, Xingyue Zhang, Chen Mingju, Yilin Wang, Xingzhong Xiong, Zhengxu Duan, Kaiyan Fu, Zhihao Lin

    The early diagnosis and precise staging of disease severity in Parkinson's disease (PD) represent significant challenges in current clinical practice, primarily limited by the subjectivity and instability of traditional assessment methods. In recent years, deep learning methods based on multimodal neuroimaging have offered a new avenue for objective and automated assessment; however, they still face key bottlenecks such as inefficient fusion of heterogeneous modalities and insufficient clinical plausibility of the models. To address these limitations, this study proposes an innovative two-stage multimodal deep learning framework, DM-PAFNet, which integrates a dual-branch multi-modal classification network (DB-MCNet) and a medical prior-guided attention fusion network (MP-AFNet). The framework is centered on a progressive strategy of "diagnosis for pre-training, staging for fine-tuning," modeling PD assessment as a hierarchical knowledge transfer process. In the first stage, we construct the DB-MCNet, which introduces a dual-stream adaptive fusion (DSAF) module to efficiently integrate information from T1-weighted imaging (T1-MRI) and diffusion tensor imaging (DTI). This stage achieves high-precision PD diagnosis and yields an image feature encoder with strong generalization capabilities. In the second stage, we propose the MP-AFNet. Building upon the inherited pre-trained weights, it utilizes a gated cross-modal attention (GCA) module to fuse imaging features with clinical variables. Furthermore, it innovatively introduces a medically-informed constraint (MIC) module, which embeds clinical knowledge as a soft constraint during training to enhance the credibility and stability of the staging predictions. Experimental results on the public PPMI dataset demonstrate the outstanding performance of our framework. The first-stage DB-MCNet achieved an accuracy of 95.85% in the PD vs. healthy control classification task, with the learned imaging encoder subsequently transferred to initialize the second stage. In the more challenging Hoehn and Yahr (H&Y) staging prediction across 410 subjects (H&Y stages 0–3), the complete framework achieved an accuracy of 88.3%, significantly outperforming several mainstream baseline methods. In conclusion, by virtue of its two-stage structural design, efficient multimodal fusion mechanisms, and clinical prior-guided strategy, the DM-PAFNet provides an accurate, objective, and clinically interpretable solution for the automated diagnosis and fine-grained staging of PD.


  • Open Access

    BCA-MVSNet: Integrating BIFPN and CA for Enhanced Detail Texture in Multi-View Stereo Reconstruction

    by Ning Long, Zhengxu Duan, Xiao Hu, Mingju Chen

    The 3D point cloud generated by MVSNet has good scene integrity but lacks sensitivity to details, causing holes and non-dense areas in flat and weak-texture regions. To address this problem and enhance the point cloud information of weak-texture areas, the BCA-MVSNet network is proposed in this paper. The network integrates the Bidirectional Feature Pyramid Network (BIFPN) into the feature processing of the MVSNet backbone network to accurately extract the features of weak-texture regions. In the feature map fusion stage, the Coordinate Attention (CA) mechanism is introduced into 3DU-Net to obtain the position information on the channel dimension related to the direction, improve the detail feature extraction, optimize the depth map and improve the depth accuracy. The experimental results show that BCA-MVSNet not only improves the accuracy of detail texture reconstruction, but also effectively controls the computational overhead. In the DTU dataset, the Overall and Comp metrics of BCA-MVSNet are reduced by 10.2% and 2.6%, respectively; in the Tanksand Temples dataset, the Mean metrics of the eight scenarios are improved by 6.51%. Three scenes are shot by binocular camera, and the reconstruction quality is excellent in the weak-texture area by combining the camera parameters and the BCA-MVSNet model.


  • Open Access

    An Efficient Image Deblurring Network with a Hybrid Architecture

    by Mingju Chen, Sihang Yi, Zhongxiao Lan, Zhengxu Duan

    Blurring is one of the main degradation factors in image degradation, so image deblurring is of great interest as a fundamental problem in low-level computer vision. Because of the limited receptive field, traditional CNNs lack global fuzzy region modeling, and do not make full use of rich context information between features. Recently, a transformer-based neural network structure has performed well in natural language tasks, inspiring rapid development in the field of defuzzification. Therefore, in this paper, a hybrid architecture based on CNN and transformers is used for image deblurring. Specifically, we first extract the shallow features of the blurred images using a cross-layer feature fusion block that emphasizes the contextual information of each feature extraction layer. Secondly, an efficient transformer module for extracting deep features is designed, which fully aggregates feature information at medium and long distances using vertical and horizontal intra- and inter-strip attention layers, and a dual gating mechanism is used as a feedforward neural network, which effectively reduces redundant features. Finally, the cross-layer feature fusion block is used to complement the feature information to obtain the deblurred image. Extensive experimental results on publicly available benchmark datasets GoPro, HIDE, and the real dataset RealBlur show that the proposed method outperforms the current mainstream deblurring algorithms and recovers the edge contours and texture details of the images more clearly.


  • Open Access

    A Two-Stage Image Inpainting Technique for Old Photographs Based on Transfer Learning

    by Mingju Chen, Zhengxu Duan, Lan Li, Sihang Yi, Anle Cui

    To address the challenge of sparse old photo datasets, we apply transfer learning to image inpainting tasks. Specifically, we improve a two-stage image inpainting network that focuses on collaborative subtasks. We also design a transform module based on the cross-aggregation of windows to improve long-distance contextual information acquisition in image inpainting and enhance the integrity of images in terms of structure and texture. Our improved two-stage network has a significantly better repair performance compared to that of the current common inpainting methods. We further apply transfer learning techniques by utilizing the improved two-stage image inpainting network as the base network and decoupling the generator into a feature extractor and classifier, which consist of an encoder and a decoder, respectively. We obtain a domain-invariant feature extractor through minimax game training using source and target domain data. This feature extractor can be combined with the original encoder to restore old photo images. To verify the effectiveness of our approach, we conducted comparative experiments. Our results show that the PSNR, SSIM, and FID indexes of the model using transfer learning are 11.8%, 2.96%, and 44.4% higher than those without transfer learning, respectively. These findings suggest that applying transfer learning techniques can be an effective solution to address the challenge of sparse old photo datasets in image inpainting tasks.


  • Open Access

    A Transformer-Based Cross-Window Aggregated Attentional Image Inpainting Model

    by Mingju Chen, Tingting Liu, Xingzhong Xiong, Zhengxu Duan, Anle Cui

    To overcome the fault of convolutional networks, which can be over-smooth, blurred, or discontinuous, a novel transformer network with cross-window aggregated attention is proposed. Our network as a whole is constructed as a generative adversarial network model, and by embedding the Window Aggregation Transformer (WAT) module, we improve the information aggregation between windows without increasing the computational complexity and effectively obtain the image long-range dependencies to solve the problem that convolutional operations are limited by local feature extraction. First, the encoder extracts the multi-scale features of the image with convolution kernels of different scales; second, the feature maps of different scales are input into a WAT module to realize the aggregation between feature information and finally, these features are reconstructed by the decoder, and then, the generated image is input into the global discriminator, in which the discrimination between real and fake images is completed. It is experimentally verified that our designed Transformer window attention network is able to make the structured texture of the restored images richer and more natural when performing the restoration task of large broken or structurally complex images.


  • Open Access

    Scene Reconstruction Algorithm for Unstructured Weak-Texture Regions Based on Stereo Vision

    by Mingju Chen, Zhengxu Duan, Zhongxiao Lan, Sihang Yi

    At present, Chinese 3D reconstruction solutions using stereo cameras mainly face known, indoor, structured scenes; for the reconstruction of unstructured, larger-scale scenes with a large variety of texture information of different intensities, there are certain difficulties in ensuring accuracy and real-time processing. For the above problems, we propose a scene reconstruction method using stereo vision. Firstly, considering the influence of outdoor lighting and weather on the captured 2D images, the optimized SAD-FAST feature detection algorithm and stereo-matching strategy were employed in the stereo-matching stage to improve the overall efficiency and matching quality at this stage. Then, a homogenized feature extraction algorithm with gradient value decreasing step by step (GVDS) was used in the depth value calculation to ensure a sufficient number of feature points for strong texture information while extracting features from weak-texture areas, which greatly improved the quality and speed of unstructured scene reconstruction. We conducted experiments to validate the proposed method, and the results showed the feasibility of the proposed method and its high practical value.