Xuechen Li
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
Xuechen Li has a total of 66 co-authors. Here are the top co-authors by number of shared publications:
- Linlin Shen
Shenzhen University, Shenzhen, China
13 shared publications - Xuguang Li
Department of Stomatology Shenzhen University General Hospital Shenzhen China
6 shared publications - Xinquan Yang
Shenzhen University
5 shared publications - Yongqiang Deng
Department of Stomatology Shenzhen University General Hospital Shenzhen China
5 shared publications - Wenting Chen
Stanford University, Stanford, USA
4 shared publications - Xinquan Yang
College of Computer Science and Software Engineering Shenzhen University Shenzhen China
3 shared publications
Latest Publications
- Open Access
- 29 June 2026
- Published by MDPI AG in Algorithms
- Vol. 19 (7) , 525
Hardware-Aware Sparse QUBO Encoding for CVRPTW on Coherent Ising Machines: An LKH-Guided Variable-Compression Framework
Capacitated vehicle routing with time windows (CVRPTW) is a natural target for coherent Ising machines (CIMs), but a direct multi-vehicle arc encoding scales as and exceeds the variable budget of current CIM-compatible systems. We argue the bottleneck is encoding density, not expressiveness, and present LSQ, a hardware-aware sparse Quadratic Unconstrained Binary Optimization (QUBO) framework that decouples CVRPTW into a compact customer-to-route assignment QUBO and a classical intra-route ordering step under a soft no-wait service convention. LKH candidate edges compress the per-route edge space from to , and a per-route dynamic-penalty subroutine encodes time-window sensitivities as binary variables in a round-wise outer loop. On a six-vehicle, 51-node reference instance curated from long-term operational data, LSQ shrinks the maximum single-submission QUBO from 15,300 arc variables to 342 logical QUBO variables (∼45× compression), cuts travel time by 22.9% (74 vs. 96), and cuts route duration by 11.2% (174 vs. 196) against an OR-Tools soft-window baseline at the same fleet size. OR-Tools retains an advantage on raw time-window penalty (1600 vs. 3540) and runtime; under the scalar operational cost , OR-Tools is therefore the better single-objective solver, and the comparison is a multi-objective trade-off rather than a scalar dominance claim. Ablations confirm that the LKH prior recovers Held–Karp on a 15-customer TSP at 53 vs. 120 variables and that the dynamic-penalty encoding reduces compressed time-window loss by 15.25% at constant travel. All hardware claims refer to QUBO sizing on a Kaiwu/CIM-compatible backend, not physical CIM execution.- 29 June 2026
- Published by MDPI AG in Algorithms
- Vol. 19 (7) , 525
Open AccessHardware-Aware Sparse QUBO Encoding for CVRPTW on Coherent Ising Machines: An LKH-Guided Variable-Compression Framework
Capacitated vehicle routing with time windows (CVRPTW) is a natural target for coherent Ising machines (CIMs), but a direct multi-vehicle arc encoding scales as and exceeds the variable budget of current CIM-compatible systems. We argue the bottleneck is encoding density, not expressiveness, and present LSQ, a hardware-aware sparse Quadratic Unconstrained Binary Optimization (QUBO) framework that decouples CVRPTW into a compact customer-to-route assignment QUBO and a classical intra-route ordering step under a soft no-wait service convention. LKH candidate edges compress the per-route edge space from to , and a per-route dynamic-penalty subroutine encodes time-window sensitivities as binary variables in a round-wise outer loop. On a six-vehicle, 51-node reference instance curated from long-term operational data, LSQ shrinks the maximum single-submission QUBO from 15,300 arc variables to 342 logical QUBO variables (∼45× compression), cuts travel time by 22.9% (74 vs. 96), and cuts route duration by 11.2% (174 vs. 196) against an OR-Tools soft-window baseline at the same fleet size. OR-Tools retains an advantage on raw time-window penalty (1600 vs. 3540) and runtime; under the scalar operational cost , OR-Tools is therefore the better single-objective solver, and the comparison is a multi-objective trade-off rather than a scalar dominance claim. Ablations confirm that the LKH prior recovers Held–Karp on a 15-customer TSP at 53 vs. 120 variables and that the dynamic-penalty encoding reduces compressed time-window loss by 15.25% at constant travel. All hardware claims refer to QUBO sizing on a Kaiwu/CIM-compatible backend, not physical CIM execution. - 12 June 2026
- Published by SAGE Publications in Molecular Imaging
- Vol. 25
Self-Supervised Learning Method for 3D Detection of Lung Cancer Based on PET Imaging
Background: Positron emission tomography (PET) is one of the most effective imaging methods for detecting lung cancer, and artificial intelligence-based approaches are increasingly being applied to PET-based lung cancer detection. However, this typically requires a large amount of labeled data for training, while obtaining sufficient labeled PET imaging data remains challenging. Objective: To improve the accuracy of three-dimensional (3D) lung cancer detection in PET images and address the issue of scarce labeled data, this study aims to propose a novel self-supervised learning method based on the pseudo image generation. Methods: This method first uses a spatial tumor simulator to generate 3D images resembling real lung cancer lesions, and randomly implants them into the lung regions of PET images to construct pseudo-lung-cancer PET images. Subsequently, a restoration task based on pseudo-cancer images and normal images is designed as a self-supervised pretraining objective, using the 55 training cases to generate paired original and pseudo-lesion PET images, which are used to pretrain a Dual-Attention Hybrid Unet (DH-Unet) encoder–decoder integrated with a self-attention mechanism. Finally, the model is fine-tuned using real labeled PET data from these same 55 cases, along with 10 additional test cases, all pathologically confirmed as lung cancer, to complete the 3D lung cancer detection task. Results: Experimental results show that this method achieves significant performance in the lung cancer detection task, with an mAP@ 0.10 to 0.50 of 0.4616, which is 13.72% higher than that of the random initialization method and 11.1% higher than that of traditional self-supervised models. Conclusion: The proposed self-supervised learning framework, which combines pseudo image generation and a self-attention-equipped DH-Unet, provides an effective approach for improving lung cancer detection in PET imaging, especially in scenarios with limited labeled data.
- 12 June 2026
- Published by SAGE Publications in Molecular Imaging
- Vol. 25
Self-Supervised Learning Method for 3D Detection of Lung Cancer Based on PET Imaging
Background: Positron emission tomography (PET) is one of the most effective imaging methods for detecting lung cancer, and artificial intelligence-based approaches are increasingly being applied to PET-based lung cancer detection. However, this typically requires a large amount of labeled data for training, while obtaining sufficient labeled PET imaging data remains challenging. Objective: To improve the accuracy of three-dimensional (3D) lung cancer detection in PET images and address the issue of scarce labeled data, this study aims to propose a novel self-supervised learning method based on the pseudo image generation. Methods: This method first uses a spatial tumor simulator to generate 3D images resembling real lung cancer lesions, and randomly implants them into the lung regions of PET images to construct pseudo-lung-cancer PET images. Subsequently, a restoration task based on pseudo-cancer images and normal images is designed as a self-supervised pretraining objective, using the 55 training cases to generate paired original and pseudo-lesion PET images, which are used to pretrain a Dual-Attention Hybrid Unet (DH-Unet) encoder–decoder integrated with a self-attention mechanism. Finally, the model is fine-tuned using real labeled PET data from these same 55 cases, along with 10 additional test cases, all pathologically confirmed as lung cancer, to complete the 3D lung cancer detection task. Results: Experimental results show that this method achieves significant performance in the lung cancer detection task, with an mAP@ 0.10 to 0.50 of 0.4616, which is 13.72% higher than that of the random initialization method and 11.1% higher than that of traditional self-supervised models. Conclusion: The proposed self-supervised learning framework, which combines pseudo image generation and a self-attention-equipped DH-Unet, provides an effective approach for improving lung cancer detection in PET imaging, especially in scenarios with limited labeled data.
- 24 March 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Pattern Recognition and Artificial Intelligence
- Vol. 40 (08)
SRP-YOLO: A Receptive Field Enhanced and Partial Convolution Fusion Network for Small Object Detection
Accurate object detection in unmanned aerial vehicle (UAV) images plays a crucial role in fields such as aviation, transportation, and agriculture. However, UAV images often contain a high proportion of small objects, and the limited resources of UAV platforms make it challenging for existing small object detection algorithms to balance detection performance and resource consumption. To address these issues, a lightweight small object detection algorithm called SRP-YOLO is proposed. Compared to the YOLOv8 network, the following improvements are made: first, a high-resolution detection head is designed; second, a receptive field attention convolution module tailored for small UAV objects is introduced to replace the standard convolution module, enhancing the detection capability for small objects and significantly improving their feature representation; Third, partial convolution is combined with C2f to replace some of the standard convolutions, fully leveraging the high-resolution information from shallow features. Finally, a normalized Gaussian Wasserstein distance (NWD) metric is introduced to reduce the sensitivity of intersection over union (IoU) to minor positional deviations of small objects. On VisDrone-DET2019, SRP-YOLO reduces parameters while improving [ ] by 7.9%. Generalization tests on TinypersonV2 and RAIVD confirm significant gains in small object detection.
- 24 March 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Pattern Recognition and Artificial Intelligence
- Vol. 40 (08)
SRP-YOLO: A Receptive Field Enhanced and Partial Convolution Fusion Network for Small Object Detection
Accurate object detection in unmanned aerial vehicle (UAV) images plays a crucial role in fields such as aviation, transportation, and agriculture. However, UAV images often contain a high proportion of small objects, and the limited resources of UAV platforms make it challenging for existing small object detection algorithms to balance detection performance and resource consumption. To address these issues, a lightweight small object detection algorithm called SRP-YOLO is proposed. Compared to the YOLOv8 network, the following improvements are made: first, a high-resolution detection head is designed; second, a receptive field attention convolution module tailored for small UAV objects is introduced to replace the standard convolution module, enhancing the detection capability for small objects and significantly improving their feature representation; Third, partial convolution is combined with C2f to replace some of the standard convolutions, fully leveraging the high-resolution information from shallow features. Finally, a normalized Gaussian Wasserstein distance (NWD) metric is introduced to reduce the sensitivity of intersection over union (IoU) to minor positional deviations of small objects. On VisDrone-DET2019, SRP-YOLO reduces parameters while improving [ ] by 7.9%. Generalization tests on TinypersonV2 and RAIVD confirm significant gains in small object detection.
- 7 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (4) , 3307-3320
ACGM: Attribute-Centric Graph Modeling Network for Concurrent Missing Tabular Data Imputation and COVID-19 Prognosis
COVID-19 prognosis using clinical tabular data faces significant challenges due to missing values and class imbalance issues. Existing methods often overlook the complex high-order interrelationship among clinical attributes and struggle with training stability on imbalanced datasets. We propose ACGM, an attribute-centric graph modeling network that simultaneously addresses missing data imputation and COVID-19 prognosis. ACGM consists of three key modules: an attributes preprocessing module (APM) for coarse-grained imputation initialization, a graph-enhanced attributes imputation module (GEAIM) that models high-order inter-attribute relationships through graph structures, and a graph-enhanced disease prognosis module (GEDPM) that leverages these complex attribute interactions for final prediction. GEAIM and GEDPM employ a mean-teacher strategy with attributes graph matching to preserve high-order relationships, enhance training stability, and maintain structural integrity of attribute interactions. Extensive experiments are conducted on four public COVID-19 tabular datasets, demonstrating the superiority of our ACGM over existing methods. Through comprehensive interpretability analysis, we identify that attributes such as LDH, Difficulty In Breathing, and SaO2 significantly impact COVID-19 prognosis, aligning well with clinical insights and radiologist assessments.
- 7 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (4) , 3307-3320
ACGM: Attribute-Centric Graph Modeling Network for Concurrent Missing Tabular Data Imputation and COVID-19 Prognosis
COVID-19 prognosis using clinical tabular data faces significant challenges due to missing values and class imbalance issues. Existing methods often overlook the complex high-order interrelationship among clinical attributes and struggle with training stability on imbalanced datasets. We propose ACGM, an attribute-centric graph modeling network that simultaneously addresses missing data imputation and COVID-19 prognosis. ACGM consists of three key modules: an attributes preprocessing module (APM) for coarse-grained imputation initialization, a graph-enhanced attributes imputation module (GEAIM) that models high-order inter-attribute relationships through graph structures, and a graph-enhanced disease prognosis module (GEDPM) that leverages these complex attribute interactions for final prediction. GEAIM and GEDPM employ a mean-teacher strategy with attributes graph matching to preserve high-order relationships, enhance training stability, and maintain structural integrity of attribute interactions. Extensive experiments are conducted on four public COVID-19 tabular datasets, demonstrating the superiority of our ACGM over existing methods. Through comprehensive interpretability analysis, we identify that attributes such as LDH, Difficulty In Breathing, and SaO2 significantly impact COVID-19 prognosis, aligning well with clinical insights and radiologist assessments.
- Open Access
- 2 January 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (1) , 361
sRrsR-Net: A New Low-Light Image Enhancement Network via Raw Image Reconstruction
Most existing low-light image enhancement (LIE) methods are primarily designed for human-vision-friendly image formats, such as sRGB, due to their convenient storage and smaller file sizes. In addition, raw images provide greater detail and a wider dynamic range, which makes them more suitable for LIE tasks. Despite these advantages, raw images, the original format captured by cameras, are larger and less accessible and are hard to use in methods of LIE with mobile devices. In order to leverage both the advantages of sRGB and raw domains while avoiding the direct use of raw images as training data, this paper introduces sRrsR-Net, a novel framework with the image translation process of sRGB–raw–sRGB for LIE task. In our approach, firstly, the RGB-to-iRGB module is designed to convert sRGB images into intermediate RGB feature maps. Then, with these intermediate feature maps, to bridge the domain gap between sRGB and raw pixels, the RAWFormer module is proposed to employ global attention to effectively align features between the two domains to generate reconstructed raw images. For enhancing the raw images and restoring them back to normal-light sRGB, unlike traditional Image Signal Processing (ISP) pipelines, which are often bulky and integrate numerous processing steps, we propose the RRAW-to-sRGB module. This module simplifies the process by focusing only on color correction and white balance, while still delivering competitive results. Extensive experiments on four benchmark datasets referring to both domains demonstrate the effectiveness of our approach.
- 2 January 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (1) , 361
Open AccesssRrsR-Net: A New Low-Light Image Enhancement Network via Raw Image Reconstruction
Most existing low-light image enhancement (LIE) methods are primarily designed for human-vision-friendly image formats, such as sRGB, due to their convenient storage and smaller file sizes. In addition, raw images provide greater detail and a wider dynamic range, which makes them more suitable for LIE tasks. Despite these advantages, raw images, the original format captured by cameras, are larger and less accessible and are hard to use in methods of LIE with mobile devices. In order to leverage both the advantages of sRGB and raw domains while avoiding the direct use of raw images as training data, this paper introduces sRrsR-Net, a novel framework with the image translation process of sRGB–raw–sRGB for LIE task. In our approach, firstly, the RGB-to-iRGB module is designed to convert sRGB images into intermediate RGB feature maps. Then, with these intermediate feature maps, to bridge the domain gap between sRGB and raw pixels, the RAWFormer module is proposed to employ global attention to effectively align features between the two domains to generate reconstructed raw images. For enhancing the raw images and restoring them back to normal-light sRGB, unlike traditional Image Signal Processing (ISP) pipelines, which are often bulky and integrate numerous processing steps, we propose the RRAW-to-sRGB module. This module simplifies the process by focusing only on color correction and white balance, while still delivering competitive results. Extensive experiments on four benchmark datasets referring to both domains demonstrate the effectiveness of our approach.
- 24 December 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 37 (6) , 4651-4661
MLPFormer: MLP-integrated transformer for colorectal histopathology whole slide image segmentation
- 24 December 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 37 (6) , 4651-4661
MLPFormer: MLP-integrated transformer for colorectal histopathology whole slide image segmentation
- 2 May 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 252
SCPMan: Shape context and prior constrained multi-scale attention network for pancreatic segmentation
- 2 May 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 252
SCPMan: Shape context and prior constrained multi-scale attention network for pancreatic segmentation
- 18 February 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (12) , 6643-6658
ImplantFormer: vision transformer-based implant position regression using dental CBCT data
- 18 February 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (12) , 6643-6658
ImplantFormer: vision transformer-based implant position regression using dental CBCT data
- Open Access
SCPMan: Shape Context and Prior Constrained Multi-scale Attention Network for Pancreatic Segmentation
Due to the poor prognosis of Pancreatic cancer, accurate early detection and segmentation are critical for improving treatment outcomes. However, pancreatic segmentation is challenged by blurred boundaries, high shape variability, and class imbalance. To tackle these problems, we propose a multiscale attention network with shape context and prior constraint for robust pancreas segmentation. Specifically, we proposed a Multi-scale Feature Extraction Module (MFE) and a Mixed-scale Attention Integration Module (MAI) to address unclear pancreas boundaries. Furthermore, a Shape Context Memory (SCM) module is introduced to jointly model semantics across scales and pancreatic shape. Active Shape Model (ASM) is further used to model the shape priors. Experiments on NIH and MSD datasets demonstrate the efficacy of our model, which improves the state-of-the-art Dice Score for 1.01% and 1.03% respectively. Our architecture provides robust segmentation performance, against the blurry boundaries, and variations in scale and shape of pancreas.
Open AccessSCPMan: Shape Context and Prior Constrained Multi-scale Attention Network for Pancreatic Segmentation
Due to the poor prognosis of Pancreatic cancer, accurate early detection and segmentation are critical for improving treatment outcomes. However, pancreatic segmentation is challenged by blurred boundaries, high shape variability, and class imbalance. To tackle these problems, we propose a multiscale attention network with shape context and prior constraint for robust pancreas segmentation. Specifically, we proposed a Multi-scale Feature Extraction Module (MFE) and a Mixed-scale Attention Integration Module (MAI) to address unclear pancreas boundaries. Furthermore, a Shape Context Memory (SCM) module is introduced to jointly model semantics across scales and pancreatic shape. Active Shape Model (ASM) is further used to model the shape priors. Experiments on NIH and MSD datasets demonstrate the efficacy of our model, which improves the state-of-the-art Dice Score for 1.01% and 1.03% respectively. Our architecture provides robust segmentation performance, against the blurry boundaries, and variations in scale and shape of pancreas.
- 1 October 2023
- Published by Springer Nature
- p. 317-326
TCEIP: Text Condition Embedded Regression Network for Dental Implant Position Prediction
- 1 October 2023
- Published by Springer Nature
- p. 317-326
TCEIP: Text Condition Embedded Regression Network for Dental Implant Position Prediction