Lei Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Lei Yu has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Shenhai Zheng
Chongqing University of Posts and Telecommunications
3 shared publications - Weisheng Li
Chongqing University of Posts and Telecommunications
3 shared publications Chaohui Yang
Department of Biochemistry and Molecular Biology, College of Basic Medical Sciences, Third Military Medical University, Chongqing, China
1 shared publication- Donghuang Hong
Department of Critical Care Medicine, Shengli Clinical Medical College of Fujian Medical University, Fujian Provincial Hospital, Fuzhou, China
1 shared publication Haiguo Zhao
1 shared publicationJianfei Li
1 shared publication
Latest Publications
- 20 January 2025
- Published by Elsevier in Pattern Recognition
- Vol. 162
Weak-prior dual cognitive attention lightweight network for abdominal multi-organ segmentation
- 20 January 2025
- Published by Elsevier in Pattern Recognition
- Vol. 162
Weak-prior dual cognitive attention lightweight network for abdominal multi-organ segmentation
- Top Cited Papers
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (4) , 1836-1852
Asymmetric Adaptive Heterogeneous Network for Multi-Modality Medical Image Segmentation
Existing studies of multi-modality medical image segmentation tend to aggregate all modalities without discrimination and employ multiple symmetric encoders or decoders for feature extraction and fusion. They often overlook the different contributions to visual representation and intelligent decisions among multi-modality images. Motivated by this discovery, this paper proposes an asymmetric adaptive heterogeneous network for multi-modality image feature extraction with modality discrimination and adaptive fusion. For feature extraction, it uses a heterogeneous two-stream asymmetric feature-bridging network to extract complementary features from auxiliary multi-modality and leading single-modality images, respectively. For feature adaptive fusion, the proposed Transformer-CNN Feature Alignment and Fusion (T-CFAF) module enhances the leading single-modality information, and the Cross-Modality Heterogeneous Graph Fusion (CMHGF) module further fuses multi-modality features at a high-level semantic layer adaptively. Comparative evaluation with ten segmentation models on six datasets demonstrates significant efficiency gains as well as highly competitive segmentation accuracy. (Our code is publicly available at https://github.com/joker-527/AAHN).
- 20 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 44 (4) , 1836-1852
Top Cited PapersAsymmetric Adaptive Heterogeneous Network for Multi-Modality Medical Image Segmentation
Existing studies of multi-modality medical image segmentation tend to aggregate all modalities without discrimination and employ multiple symmetric encoders or decoders for feature extraction and fusion. They often overlook the different contributions to visual representation and intelligent decisions among multi-modality images. Motivated by this discovery, this paper proposes an asymmetric adaptive heterogeneous network for multi-modality image feature extraction with modality discrimination and adaptive fusion. For feature extraction, it uses a heterogeneous two-stream asymmetric feature-bridging network to extract complementary features from auxiliary multi-modality and leading single-modality images, respectively. For feature adaptive fusion, the proposed Transformer-CNN Feature Alignment and Fusion (T-CFAF) module enhances the leading single-modality information, and the Cross-Modality Heterogeneous Graph Fusion (CMHGF) module further fuses multi-modality features at a high-level semantic layer adaptively. Comparative evaluation with ten segmentation models on six datasets demonstrates significant efficiency gains as well as highly competitive segmentation accuracy. (Our code is publicly available at https://github.com/joker-527/AAHN).
- Open Access
- 13 June 2024
- Published by Wolters Kluwer Health in Clinical and Translational Gastroenterology
- Vol. 15 (7) , e00726
Metagenomic Next-Generation Sequencing-Based Fine-Needle Aspiration in Patients With Suspected Infected Pancreatic Necrosis
INTRODUCTION: Fine-needle aspiration (FNA) is no longer recommended for diagnosing infected pancreatic necrosis (IPN) due to a high false-negative rate. Metagenomic next-generation sequencing (mNGS) is a valuable tool for identifying potential pathogens. We hypothesized that adding mNGS to the standard FNA procedure may increase diagnostic accuracy. METHODS: This is a prospective, single-arm feasibility study enrolling patients with acute necrotizing pancreatitis complicated by suspected IPN. Computed tomography-guided FNA was performed immediately after enrollment, and the drainage samples were subjected to culture and mNGS assays simultaneously. Confirmatory IPN within the following week of the index FNA procedure was the reference standard. The diagnostic performance of FNA-mNGS and the impact of mNGS results on treatment were evaluated. Historical controls were used for comparison of clinical outcomes. RESULTS: There was no significant difference between mNGS and culture in the positive rate (75% vs 70%, P = 0.723). The accuracy of FNA-mNGS was 80.0%, with a sensitivity of 82.35%, specificity of 66.67%, positive predictive value of 93.3%, and negative predictive value of 40.0%. The results of the mNGS led to treatment change in 16 of 20 patients (80%), including implementing percutaneous catheter drainage (n = 7), expanding antibiotic coverage (n = 2), percutaneous catheter drainage and expanding coverage (n = 4), narrowing antibiotic coverage (n = 1), and discontinuation of antibiotics (n = 2). The FNA-mNGS approach was not associated with improved clinical outcomes compared with the historical control group. DISCUSSION: The addition of mNGS to standard FNA has comparable diagnostic accuracy with culture-based FNA and may not be associated with improved clinical outcomes.
- 13 June 2024
- Published by Wolters Kluwer Health in Clinical and Translational Gastroenterology
- Vol. 15 (7) , e00726
Open AccessMetagenomic Next-Generation Sequencing-Based Fine-Needle Aspiration in Patients With Suspected Infected Pancreatic Necrosis
INTRODUCTION: Fine-needle aspiration (FNA) is no longer recommended for diagnosing infected pancreatic necrosis (IPN) due to a high false-negative rate. Metagenomic next-generation sequencing (mNGS) is a valuable tool for identifying potential pathogens. We hypothesized that adding mNGS to the standard FNA procedure may increase diagnostic accuracy. METHODS: This is a prospective, single-arm feasibility study enrolling patients with acute necrotizing pancreatitis complicated by suspected IPN. Computed tomography-guided FNA was performed immediately after enrollment, and the drainage samples were subjected to culture and mNGS assays simultaneously. Confirmatory IPN within the following week of the index FNA procedure was the reference standard. The diagnostic performance of FNA-mNGS and the impact of mNGS results on treatment were evaluated. Historical controls were used for comparison of clinical outcomes. RESULTS: There was no significant difference between mNGS and culture in the positive rate (75% vs 70%, P = 0.723). The accuracy of FNA-mNGS was 80.0%, with a sensitivity of 82.35%, specificity of 66.67%, positive predictive value of 93.3%, and negative predictive value of 40.0%. The results of the mNGS led to treatment change in 16 of 20 patients (80%), including implementing percutaneous catheter drainage (n = 7), expanding antibiotic coverage (n = 2), percutaneous catheter drainage and expanding coverage (n = 4), narrowing antibiotic coverage (n = 1), and discontinuation of antibiotics (n = 2). The FNA-mNGS approach was not associated with improved clinical outcomes compared with the historical control group. DISCUSSION: The addition of mNGS to standard FNA has comparable diagnostic accuracy with culture-based FNA and may not be associated with improved clinical outcomes.