Junde Wu
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
Research Subjects
Junde Wu has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Yanwu Xu
South China University of Technology
3 shared publications - Yu Zhang
National University of Defense Technology, Hefei, China
2 shared publications - Yueming Jin
National University of Singapore
2 shared publications - Binghong Wu
Tencent Hunyuan, Shanghai, China
1 shared publication Bingyuan Liu
École de technologie supérieure, Montreal, Montreal, Canada
1 shared publicationChubin Ou
Weizhi Medical Technology Company, Suzhou, China
1 shared publication
Latest Publications
- 6 November 2025
- Published by Elsevier in Information Fusion
- Vol. 127
Balancing light and shadow: Multi-modal image exposure restoration via frequency-phase-driven RGB-IR integration
- 6 November 2025
- Published by Elsevier in Information Fusion
- Vol. 127
Balancing light and shadow: Multi-modal image exposure restoration via frequency-phase-driven RGB-IR integration
- Top Cited PapersOpen Access
- 19 March 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 102, 103547
Medical SAM adapter: Adapting segment anything model for medical image segmentation
- 19 March 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 102, 103547
Top Cited PapersOpen AccessMedical SAM adapter: Adapting segment anything model for medical image segmentation
- 26 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 43 (9) , 3331-3342
Calibrate the Inter-Observer Segmentation Uncertainty via Diagnosis-First Principle
Many of the tissues/lesions in the medical images may be ambiguous. Therefore, medical segmentation is typically annotated by a group of clinical experts to mitigate personal bias. A common solution to fuse different annotations is the majority vote, e.g., taking the average of multiple labels. However, such a strategy ignores the difference between the grader expertness. Inspired by the observation that medical image segmentation is usually used to assist the disease diagnosis in clinical practice, we propose the diagnosis-first principle, which is to take disease diagnosis as the criterion to calibrate the inter-observer segmentation uncertainty. Following this idea, a framework named Diagnosis-First segmentation Framework (DiFF) is proposed. Specifically, DiFF will first learn to fuse the multi-rater segmentation labels to a single ground-truth which could maximize the disease diagnosis performance. We dubbed the fused ground-truth as Diagnosis-First Ground-truth (DF-GT). Then, the Take and Give Model (T&G Model) to segment DF-GT from the raw image is proposed. With the T&G Model, DiFF can learn the segmentation with the calibrated uncertainty that facilitate the disease diagnosis. We verify the effectiveness of DiFF on three different medical segmentation tasks: optic-disc/optic-cup (OD/OC) segmentation on fundus images, thyroid nodule segmentation on ultrasound images, and skin lesion segmentation on dermoscopic images. Experimental results show that the proposed DiFF can effectively calibrate the segmentation uncertainty, and thus significantly facilitate the corresponding disease diagnosis, which outperforms previous state-of-the-art multi-rater learning methods.
- 26 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 43 (9) , 3331-3342
Calibrate the Inter-Observer Segmentation Uncertainty via Diagnosis-First Principle
Many of the tissues/lesions in the medical images may be ambiguous. Therefore, medical segmentation is typically annotated by a group of clinical experts to mitigate personal bias. A common solution to fuse different annotations is the majority vote, e.g., taking the average of multiple labels. However, such a strategy ignores the difference between the grader expertness. Inspired by the observation that medical image segmentation is usually used to assist the disease diagnosis in clinical practice, we propose the diagnosis-first principle, which is to take disease diagnosis as the criterion to calibrate the inter-observer segmentation uncertainty. Following this idea, a framework named Diagnosis-First segmentation Framework (DiFF) is proposed. Specifically, DiFF will first learn to fuse the multi-rater segmentation labels to a single ground-truth which could maximize the disease diagnosis performance. We dubbed the fused ground-truth as Diagnosis-First Ground-truth (DF-GT). Then, the Take and Give Model (T&G Model) to segment DF-GT from the raw image is proposed. With the T&G Model, DiFF can learn the segmentation with the calibrated uncertainty that facilitate the disease diagnosis. We verify the effectiveness of DiFF on three different medical segmentation tasks: optic-disc/optic-cup (OD/OC) segmentation on fundus images, thyroid nodule segmentation on ultrasound images, and skin lesion segmentation on dermoscopic images. Experimental results show that the proposed DiFF can effectively calibrate the segmentation uncertainty, and thus significantly facilitate the corresponding disease diagnosis, which outperforms previous state-of-the-art multi-rater learning methods.
- Top Cited Papers
- 18 September 2023
- Published by Elsevier in Medical Image Analysis
- Vol. 90, 102938
GAMMA challenge: Glaucoma grAding from Multi-Modality imAges
- 18 September 2023
- Published by Elsevier in Medical Image Analysis
- Vol. 90, 102938
Top Cited PapersGAMMA challenge: Glaucoma grAding from Multi-Modality imAges
- 1 January 2023
- Published by Optica Publishing Group
- p. STh4H.3
An Efficient Deep Learning-Based Electromagnetic Response Prediction Model for Coding Grid Meta-atoms
We proposed an efficient deep learning-based ElectroMagnetic (EM) response prediction model for coding grid meta-atoms. Prediction errors are less than 3° and 1.5% in phase and reflectivity.
- 1 January 2023
- Published by Optica Publishing Group
- p. STh4H.3
An Efficient Deep Learning-Based Electromagnetic Response Prediction Model for Coding Grid Meta-atoms
We proposed an efficient deep learning-based ElectroMagnetic (EM) response prediction model for coding grid meta-atoms. Prediction errors are less than 3° and 1.5% in phase and reflectivity.
- Open Access
Progressive Hard-case Mining across Pyramid Levels for Object Detection
In object detection, multi-level prediction (e.g., FPN) and reweighting skills (e.g., focal loss) have drastically improved one-stage detector performance. However, the synergy between these two techniques is not fully explored in a unified framework. We find that, during training, the one-stage detector's optimization is not only restricted to the static hard-case mining loss (gradient drift) but also suffered from the diverse positive samples' proportions split by different pyramid levels (level discrepancy). Under this concern, we propose Hierarchical Progressive Focus (HPF) consisting of two key designs: 1) progressive focus, a more flexible hard-case mining setting calculated adaptive to the convergence progress, 2) hierarchical sampling, automatically generating a set of progressive focus for level-specific target optimization. Based on focal loss with ATSS-R50, our approach achieves 40.5 AP, surpassing the state-of-the-art QFL (Quality Focal Loss, 39.9 AP) and VFL (Varifocal Loss, 40.1 AP). Our best model achieves 55.1 AP on COCO test-dev, obtaining excellent results with only a typical training setting. Moreover, as a plug-and-play scheme, HPF can cooperate well with recent advances, providing a stable performance improvement on nine mainstream detectors.
Open AccessProgressive Hard-case Mining across Pyramid Levels for Object Detection
In object detection, multi-level prediction (e.g., FPN) and reweighting skills (e.g., focal loss) have drastically improved one-stage detector performance. However, the synergy between these two techniques is not fully explored in a unified framework. We find that, during training, the one-stage detector's optimization is not only restricted to the static hard-case mining loss (gradient drift) but also suffered from the diverse positive samples' proportions split by different pyramid levels (level discrepancy). Under this concern, we propose Hierarchical Progressive Focus (HPF) consisting of two key designs: 1) progressive focus, a more flexible hard-case mining setting calculated adaptive to the convergence progress, 2) hierarchical sampling, automatically generating a set of progressive focus for level-specific target optimization. Based on focal loss with ATSS-R50, our approach achieves 40.5 AP, surpassing the state-of-the-art QFL (Quality Focal Loss, 39.9 AP) and VFL (Varifocal Loss, 40.1 AP). Our best model achieves 55.1 AP on COCO test-dev, obtaining excellent results with only a typical training setting. Moreover, as a plug-and-play scheme, HPF can cooperate well with recent advances, providing a stable performance improvement on nine mainstream detectors.