Xiaoyu 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
Xiaoyu Li has a total of 115 co-authors. Here are the top co-authors by number of shared publications:
- Guowu Yang
University of Electronic Science and Technology of China
7 shared publications Shan Yang
7 shared publications- Lianhui Yu
The School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
5 shared publications - Zhigang Wang
School of Electronic Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China
5 shared publications - Geng Chen
The School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan, China
4 shared publications Geng Chen
3 shared publications
Latest Publications
- Open Access
- 10 November 2026
- Published by MDPI AG in Computation
- Vol. 14 (11) , 284
QFCNN: A Dual-Branch Hybrid Quantum–Classical Convolutional Neural Network Based on the Quantum Fourier Transform for Image Classification
Image classification is a fundamental task in computer vision. Although classical convolutional neural networks are particularly effective at extracting local spatial features, they are inherently limited in their ability to capture global frequency-domain information. To address this limitation, this paper introduces a hybrid model that couples a classical convolutional feature extraction branch with a quantum feature extraction branch built upon the quantum Fourier transform. The quantum branch maps the input into a low-dimensional representation and processes it through a sequence of quantum encoding, the quantum Fourier transform, a variational quantum circuit, and the inverse quantum Fourier transform; the resulting quantum features are concatenated with the classical features and jointly used for the final classification. Ablation studies quantify the individual contribution of each quantum component, and systematic experiments on the MNIST, Fashion-MNIST, and CIFAR-10 benchmark datasets, conducted over multiple independent random seeds on a strictly held-out test set, show that the complete architecture consistently outperforms its purely classical counterparts. All results and comparative analyses are presented in the experimental section.
- 10 November 2026
- Published by MDPI AG in Computation
- Vol. 14 (11) , 284
Open AccessQFCNN: A Dual-Branch Hybrid Quantum–Classical Convolutional Neural Network Based on the Quantum Fourier Transform for Image Classification
Image classification is a fundamental task in computer vision. Although classical convolutional neural networks are particularly effective at extracting local spatial features, they are inherently limited in their ability to capture global frequency-domain information. To address this limitation, this paper introduces a hybrid model that couples a classical convolutional feature extraction branch with a quantum feature extraction branch built upon the quantum Fourier transform. The quantum branch maps the input into a low-dimensional representation and processes it through a sequence of quantum encoding, the quantum Fourier transform, a variational quantum circuit, and the inverse quantum Fourier transform; the resulting quantum features are concatenated with the classical features and jointly used for the final classification. Ablation studies quantify the individual contribution of each quantum component, and systematic experiments on the MNIST, Fashion-MNIST, and CIFAR-10 benchmark datasets, conducted over multiple independent random seeds on a strictly held-out test set, show that the complete architecture consistently outperforms its purely classical counterparts. All results and comparative analyses are presented in the experimental section.
- 28 September 2026
- Published by Springer Nature in Quantum Information Processing
- Vol. 25 (10) , 323
SymLQA: topology-aware local quantum annealing for planar face-flux HUBO problems
- 28 September 2026
- Published by Springer Nature in Quantum Information Processing
- Vol. 25 (10) , 323
SymLQA: topology-aware local quantum annealing for planar face-flux HUBO problems
- Open Access
QG-WRN: A Quantum-Enhanced Graph Convolutional Wide Residual Network for ASD Diagnosis via Neuroimaging Sensing Technology
The pathological mechanism of autism spectrum disorder (ASD) exhibits dual heterogeneity: abnormal local energy metabolism and brain-wide high-order topological failure. To synergistically characterize these complex signals captured by advanced neuroimaging sensors, we propose the Quantum-Enhanced Graph Convolutional Wide Residual Network (QG-WRN), a modality-specific, decoupled parallel dual-stream architecture. In the classical branch, to accurately capture the spatial distribution of local metabolic abnormalities, we employ a wide residual network (WRN) to extract amplitude of low-frequency fluctuation (ALFF) features, leveraging its expanded feature channels to effectively mine regional neurodynamic properties. Furthermore, to overcome the representational bottlenecks of classical linear operators in parsing hidden, long-range network connections, we introduce quantum computing, exploiting its exponentially expansive state space and intrinsic low-parameter regularization mechanism. Guided by these properties, the quantum branch utilizes a variational quantum graph convolutional (QGCN) module—featuring a trainable circular encoding strategy and a hardware-efficient 4-qubit configuration—with a 2-layer nested message passing structure to process the functional connectivity (FC) matrix, harnessing quantum interference in Hilbert space to parse complex topology while effectively mitigating overfitting on small-sample medical data. A unified training scheme achieves full-dimensional fusion of node activity and topology. Achieving 68.49% accuracy, our method outperforms 10 classic and recent new baselines, providing a powerful computational intelligence tool for sensor-based ASD clinical diagnosis. Furthermore, interpretability analysis successfully maps core disease hubs to standard AAL116 atlas coordinates, providing a powerful tool for computationally aided ASD diagnosis.
Open AccessQG-WRN: A Quantum-Enhanced Graph Convolutional Wide Residual Network for ASD Diagnosis via Neuroimaging Sensing Technology
The pathological mechanism of autism spectrum disorder (ASD) exhibits dual heterogeneity: abnormal local energy metabolism and brain-wide high-order topological failure. To synergistically characterize these complex signals captured by advanced neuroimaging sensors, we propose the Quantum-Enhanced Graph Convolutional Wide Residual Network (QG-WRN), a modality-specific, decoupled parallel dual-stream architecture. In the classical branch, to accurately capture the spatial distribution of local metabolic abnormalities, we employ a wide residual network (WRN) to extract amplitude of low-frequency fluctuation (ALFF) features, leveraging its expanded feature channels to effectively mine regional neurodynamic properties. Furthermore, to overcome the representational bottlenecks of classical linear operators in parsing hidden, long-range network connections, we introduce quantum computing, exploiting its exponentially expansive state space and intrinsic low-parameter regularization mechanism. Guided by these properties, the quantum branch utilizes a variational quantum graph convolutional (QGCN) module—featuring a trainable circular encoding strategy and a hardware-efficient 4-qubit configuration—with a 2-layer nested message passing structure to process the functional connectivity (FC) matrix, harnessing quantum interference in Hilbert space to parse complex topology while effectively mitigating overfitting on small-sample medical data. A unified training scheme achieves full-dimensional fusion of node activity and topology. Achieving 68.49% accuracy, our method outperforms 10 classic and recent new baselines, providing a powerful computational intelligence tool for sensor-based ASD clinical diagnosis. Furthermore, interpretability analysis successfully maps core disease hubs to standard AAL116 atlas coordinates, providing a powerful tool for computationally aided ASD diagnosis.
- 17 April 2026
- Published by Elsevier in Applied Radiation and Isotopes
- Vol. 233, 112635
Automatic segmentation of target volume of breast radiotherapy on CT using MSR-UNet: A multicenter study
- 17 April 2026
- Published by Elsevier in Applied Radiation and Isotopes
- Vol. 233, 112635
Automatic segmentation of target volume of breast radiotherapy on CT using MSR-UNet: A multicenter study
- Open Access
- 27 March 2026
- Published by Association for Computing Machinery (ACM)
- p. 167-174
A Structure-Aware Homotopy Continuation Method for Finding Bifurcation Points in Ising Optimization
Combinatorial optimization problems like Ising and HUBO models are pivotal yet often solved by opaque heuristic algorithms. We propose a novel homotopy-based quantum-inspired algorithm to explicitly calculate and visualize bifurcation points in the optimization landscape. Utilizing arc-length continuation and Jacobian singularity detection, our method traces solution paths and switches branches to explore lower-energy regions. We demonstrate this framework on large-scale HUBO instances (> 1000 nodes), visualizing solution evolution via bifurcation diagrams and Principal Component Analysis (PCA) trajectories. Results identify critical phase transitions around the homotopy parameter p ≈ 0.18 and reveal U-shaped path topologies indicative of bypassing energy barriers. This approach not only ensures high-performance solving but also serves as a powerful analytical tool for unveiling the topological structure of complex energy landscapes.
- 27 March 2026
- Published by Association for Computing Machinery (ACM)
- p. 167-174
Open AccessA Structure-Aware Homotopy Continuation Method for Finding Bifurcation Points in Ising Optimization
Combinatorial optimization problems like Ising and HUBO models are pivotal yet often solved by opaque heuristic algorithms. We propose a novel homotopy-based quantum-inspired algorithm to explicitly calculate and visualize bifurcation points in the optimization landscape. Utilizing arc-length continuation and Jacobian singularity detection, our method traces solution paths and switches branches to explore lower-energy regions. We demonstrate this framework on large-scale HUBO instances (> 1000 nodes), visualizing solution evolution via bifurcation diagrams and Principal Component Analysis (PCA) trajectories. Results identify critical phase transitions around the homotopy parameter p ≈ 0.18 and reveal U-shaped path topologies indicative of bypassing energy barriers. This approach not only ensures high-performance solving but also serves as a powerful analytical tool for unveiling the topological structure of complex energy landscapes.
- Open Access
- 24 March 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
A machine learning-derived biological age model for liver grafts provides a superior assessment of aging compared to chronological age in transplantation
Liver transplantation is severely restricted by shortages of donors, yet relying on chronological age (CA) for donor selection often results in potentially viable grafts being discarded. This study developed a machine learning-based framework to predict the biological age (BA) of liver grafts and evaluate its clinical utility. Data from 247 donors were analyzed to create BA models using conventional linear methods and four genetic algorithm-optimized machine learning models. These models were then validated in an independent temporal cohort of 82 donor-recipient pairs. While multiple linear regression (MLR) showed a higher correlation (r = 0.82), the gradient boosting (GB) model captured unique nonlinear biological signals. Importantly, BA derived from the GB model demonstrated superior clinical relevance, serving as an independent predictor of biliary complications (AUC = 0.80) and graft survival, whereas CA failed to predict these outcomes. Multivariable Cox regression analysis revealed that accelerated ageing, defined as a BA exceeding CA, was an independent risk factor for graft loss (adjusted hazard ratio = 4.188). This exploratory study suggests that BA captures the aging heterogeneity of liver grafts which cannot be identified by the CA of donors. Furthermore, the genetic algorithm-optimized model shows potential in predicting biliary complications and survival outcomes.
- 24 March 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessA machine learning-derived biological age model for liver grafts provides a superior assessment of aging compared to chronological age in transplantation
Liver transplantation is severely restricted by shortages of donors, yet relying on chronological age (CA) for donor selection often results in potentially viable grafts being discarded. This study developed a machine learning-based framework to predict the biological age (BA) of liver grafts and evaluate its clinical utility. Data from 247 donors were analyzed to create BA models using conventional linear methods and four genetic algorithm-optimized machine learning models. These models were then validated in an independent temporal cohort of 82 donor-recipient pairs. While multiple linear regression (MLR) showed a higher correlation (r = 0.82), the gradient boosting (GB) model captured unique nonlinear biological signals. Importantly, BA derived from the GB model demonstrated superior clinical relevance, serving as an independent predictor of biliary complications (AUC = 0.80) and graft survival, whereas CA failed to predict these outcomes. Multivariable Cox regression analysis revealed that accelerated ageing, defined as a BA exceeding CA, was an independent risk factor for graft loss (adjusted hazard ratio = 4.188). This exploratory study suggests that BA captures the aging heterogeneity of liver grafts which cannot be identified by the CA of donors. Furthermore, the genetic algorithm-optimized model shows potential in predicting biliary complications and survival outcomes.
- 13 June 2025
- Published by Elsevier in Information Sciences
- Vol. 718
Knowledge-driven quantum architecture search through filtering and focusing
- 13 June 2025
- Published by Elsevier in Information Sciences
- Vol. 718
Knowledge-driven quantum architecture search through filtering and focusing
- 22 May 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 173
MLQM: Machine learning approach for accelerating optimal qubit mapping
- 22 May 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 173
MLQM: Machine learning approach for accelerating optimal qubit mapping
- 13 May 2025
- Published by Springer Nature
- p. 221-226
On Logical Omniscience with STRF
- Open Access
HAQA: A Hardware-Guided and Fidelity-Aware Strategy for Efficient Qubit Mapping Optimization
Quantum algorithms rely on quantum computers for implementation, but the physical connectivity constraints of modern quantum processors impede the efficient realization of quantum algorithms. Qubit mapping, a critical technology for practical quantum computing applications, directly determines the execution efficiency and feasibility of algorithms on superconducting quantum processors. Existing mapping methods overlook intractable quantum hardware fidelity characteristics, reducing circuit execution quality. They also exhibit prolonged solving times or even failure to complete when handling large-scale quantum architectures, compromising efficiency. To address these challenges, we propose a novel qubit mapping method HAQA. HAQA first introduces a community-based iterative region identification strategy leveraging hardware connection topology, achieving effective dimensionality reduction of mapping space. This strategy avoids global search procedures, with complexity analysis demonstrating quadratic polynomial-level acceleration. Furthermore, HAQA implements a hardware-characteristic-based region evaluation mechanism, enabling quantitative selection of mapping regions based on fidelity metrics. This approach effectively integrates hardware fidelity information into the mapping process, enabling fidelity-aware qubit allocation. Experimental results demonstrate that HAQA significantly improves solving speed and fidelity while ensuring solution quality. When applied to state-of-the-art quantum mapping techniques Qsynth-v2 and TB-OLSQ2, HAQA achieves acceleration ratios of 632.76 and 286.87 respectively, while improving fidelity by up to 52.69% and 238.28%
Open AccessHAQA: A Hardware-Guided and Fidelity-Aware Strategy for Efficient Qubit Mapping Optimization
Quantum algorithms rely on quantum computers for implementation, but the physical connectivity constraints of modern quantum processors impede the efficient realization of quantum algorithms. Qubit mapping, a critical technology for practical quantum computing applications, directly determines the execution efficiency and feasibility of algorithms on superconducting quantum processors. Existing mapping methods overlook intractable quantum hardware fidelity characteristics, reducing circuit execution quality. They also exhibit prolonged solving times or even failure to complete when handling large-scale quantum architectures, compromising efficiency. To address these challenges, we propose a novel qubit mapping method HAQA. HAQA first introduces a community-based iterative region identification strategy leveraging hardware connection topology, achieving effective dimensionality reduction of mapping space. This strategy avoids global search procedures, with complexity analysis demonstrating quadratic polynomial-level acceleration. Furthermore, HAQA implements a hardware-characteristic-based region evaluation mechanism, enabling quantitative selection of mapping regions based on fidelity metrics. This approach effectively integrates hardware fidelity information into the mapping process, enabling fidelity-aware qubit allocation. Experimental results demonstrate that HAQA significantly improves solving speed and fidelity while ensuring solution quality. When applied to state-of-the-art quantum mapping techniques Qsynth-v2 and TB-OLSQ2, HAQA achieves acceleration ratios of 632.76 and 286.87 respectively, while improving fidelity by up to 52.69% and 238.28%