Chun Yuan
YOU?
Author Swipe
View article: Sparse Model Inversion: Efficient Inversion of Vision Transformers for Data-Free Applications
Sparse Model Inversion: Efficient Inversion of Vision Transformers for Data-Free Applications Open
Model inversion, which aims to reconstruct the original training data from pre-trained discriminative models, is especially useful when the original training data is unavailable due to privacy, usage rights, or size constraints. However, e…
View article: Composition of carotid plaques differs between Chinese and US patients: a histology study
Composition of carotid plaques differs between Chinese and US patients: a histology study Open
Background The clinical manifestations of cerebrovascular disease are known to differ between the Chinese and United States (US) populations as do the plaque features on imaging. Objectives The aim of this study was to investigate and comp…
View article: Carotid Plaque Characteristics Combined With Serum Inflammatory Biomarkers Predict Recurrent Ischemic Stroke
Carotid Plaque Characteristics Combined With Serum Inflammatory Biomarkers Predict Recurrent Ischemic Stroke Open
Objective To investigate the potential value of serum inflammatory markers combined with carotid plaque characteristics for predicting subsequent recurrent ischemic events. Methods Patients with cerebral infarction localized to the interna…
View article: DiffuSpec: Unlocking Diffusion Language Models for Speculative Decoding
DiffuSpec: Unlocking Diffusion Language Models for Speculative Decoding Open
As large language models (LLMs) scale up, accuracy improves, but the autoregressive (AR) nature of decoding increases latency since each token requires a serial forward pass. Speculative decoding addresses this by employing a fast drafter …
View article: Metabolically Abnormal Obesity and Carotid Plaque Vulnerability: A Vessel Wall MRI Study Linking Obesity Phenotypes to Atherosclerotic Instability
Metabolically Abnormal Obesity and Carotid Plaque Vulnerability: A Vessel Wall MRI Study Linking Obesity Phenotypes to Atherosclerotic Instability Open
BACKGROUND: Carotid plaque vulnerability, driven by metabolic dysfunction and obesity, is a critical determinant of ischemic stroke risk. However, the heterogeneity of obesity phenotypes—defined by metabolic health—remains underexplored in…
View article: Unified and Semantically Grounded Domain Adaptation for Medical Image Segmentation
Unified and Semantically Grounded Domain Adaptation for Medical Image Segmentation Open
Most prior unsupervised domain adaptation approaches for medical image segmentation are narrowly tailored to either the source-accessible setting, where adaptation is guided by source-target alignment, or the source-free setting, which typ…
View article: Composition of Carotid Plaques Differs Between Chinese and United States Patients: A Histology Study
Composition of Carotid Plaques Differs Between Chinese and United States Patients: A Histology Study Open
Background: The clinical manifestations of cerebrovascular disease are known to differ between the Chinese and United States (U.S.) populations as do the plaque features on imaging. Objectives: The aim of this study was to investigate and …
View article: Depth-Sequence Transformer (DST) for Segment-Specific ICA Calcification Mapping on Non-Contrast CT
Depth-Sequence Transformer (DST) for Segment-Specific ICA Calcification Mapping on Non-Contrast CT Open
While total intracranial carotid artery calcification (ICAC) volume is an established stroke biomarker, growing evidence shows this aggregate metric ignores the critical influence of plaque location, since calcification in different segmen…
View article: Composition of Carotid Plaques Differs Between Chinese and United States Patients: A Histology Study
Composition of Carotid Plaques Differs Between Chinese and United States Patients: A Histology Study Open
Background: The clinical manifestations of cerebrovascular disease are known to differ between the Chinese and United States (U.S.) populations as do the plaque features on imaging. Objectives: The aim of this study was to investigate and …
View article: SCOUT: Teaching Pre-trained Language Models to Enhance Reasoning via Flow Chain-of-Thought
SCOUT: Teaching Pre-trained Language Models to Enhance Reasoning via Flow Chain-of-Thought Open
Chain of Thought (CoT) prompting improves the reasoning performance of large language models (LLMs) by encouraging step by step thinking. However, CoT-based methods depend on intermediate reasoning steps, which limits scalability and gener…
View article: Plaque-Reporting and Data System Related to Cerebrovascular Event Risk with Mild/Moderate Stenosis: A Chinese Atherosclerosis Risk Evaluation II Study
Plaque-Reporting and Data System Related to Cerebrovascular Event Risk with Mild/Moderate Stenosis: A Chinese Atherosclerosis Risk Evaluation II Study Open
Male patients of advanced age with a smoking history were associated with an increased risk of higher Plaque-RADS scores. Plaque-RADS demonstrated the ability to stratify patients experiencing cerebrovascular events, even in cases with mil…
View article: Qualitative and quantitative reproducibility of 3D MERGE and SNAP sequences for carotid vessel wall imaging across Siemens and Philips 3T scanners
Qualitative and quantitative reproducibility of 3D MERGE and SNAP sequences for carotid vessel wall imaging across Siemens and Philips 3T scanners Open
3D MERGE and SNAP images have sufficient image quality and consistent quantitative measurements on Siemens and Philips scanners, despite lower image quality in Siemens platforms, probably due to suboptimal coil configuration or image proce…
View article: Zero Token-Driven Deep Thinking in LLMs: Unlocking the Full Potential of Existing Parameters via Cyclic Refinement
Zero Token-Driven Deep Thinking in LLMs: Unlocking the Full Potential of Existing Parameters via Cyclic Refinement Open
Resource limitations often constrain the parameter counts of Large Language Models (LLMs), hindering their performance. While existing methods employ parameter sharing to reuse the same parameter set under fixed budgets, such approaches ty…
View article: RemInD: Remembering Anatomical Variations for Interpretable Domain Adaptive Medical Image Segmentation
RemInD: Remembering Anatomical Variations for Interpretable Domain Adaptive Medical Image Segmentation Open
This work presents a novel Bayesian framework for unsupervised domain adaptation (UDA) in medical image segmentation. While prior works have explored this clinically significant task using various strategies of domain alignment, they often…
View article: GSRender: Deduplicated Occupancy Prediction via Weakly Supervised 3D Gaussian Splatting
GSRender: Deduplicated Occupancy Prediction via Weakly Supervised 3D Gaussian Splatting Open
3D occupancy perception is gaining increasing attention due to its capability to offer detailed and precise environment representations. Previous weakly-supervised NeRF methods balance efficiency and accuracy, with mIoU varying by 5-10 poi…
View article: Long-Term Carotid Plaque Progression and the Role of Intraplaque Hemorrhage: A Deep Learning-Based Analysis of Longitudinal Vessel Wall Imaging
Long-Term Carotid Plaque Progression and the Role of Intraplaque Hemorrhage: A Deep Learning-Based Analysis of Longitudinal Vessel Wall Imaging Open
Background Carotid atherosclerosis is a major contributor in the etiology of ischemic stroke. Although intraplaque hemorrhage (IPH) is known to increase stroke risk and plaque burden, its long-term effects on plaque dynamics remain unclear…
View article: MoCoLSK: Modality Conditioned High-Resolution Downscaling for Land Surface Temperature
MoCoLSK: Modality Conditioned High-Resolution Downscaling for Land Surface Temperature Open
Land Surface Temperature (LST) is a critical parameter for environmental studies, but directly obtaining high spatial resolution LST data remains challenging due to the spatio-temporal trade-off in satellite remote sensing. Guided LST down…
View article: Enhancing Logits Distillation with Plug\&Play Kendall's $τ$ Ranking Loss
Enhancing Logits Distillation with Plug\&Play Kendall's $τ$ Ranking Loss Open
Knowledge distillation typically minimizes the Kullback-Leibler (KL) divergence between teacher and student logits. However, optimizing the KL divergence can be challenging for the student and often leads to sub-optimal solutions. We furth…
View article: Efficient and Accurate 3D Thickness Measurement in Vessel Wall Imaging: Overcoming Limitations of 2D Approaches Using the Laplacian Method
Efficient and Accurate 3D Thickness Measurement in Vessel Wall Imaging: Overcoming Limitations of 2D Approaches Using the Laplacian Method Open
The clinical significance of measuring vessel wall thickness is widely acknowledged. Recent advancements have enabled high-resolution 3D scans of arteries and precise segmentation of their lumens and outer walls; however, most existing met…
View article: Learn To Learn More Precisely
Learn To Learn More Precisely Open
Meta-learning has been extensively applied in the domains of few-shot learning and fast adaptation, achieving remarkable performance. While Meta-learning methods like Model-Agnostic Meta-Learning (MAML) and its variants provide a good set …
View article: Estimating Flow Direction of Circle of Willis Using Dynamic Arterial Spin-Labeling MR Angiography
Estimating Flow Direction of Circle of Willis Using Dynamic Arterial Spin-Labeling MR Angiography Open
Using ASL dMRA and the developed image analysis method to estimate the flow direction in COW is feasible. This study provides a new method to assess the hemodynamics of the COW, which could be useful for the diagnosis and study of cerebrov…
View article: Personalized Intervention Based on Early Detection of Atherosclerosis
Personalized Intervention Based on Early Detection of Atherosclerosis Open
View article: Blind Face Restoration under Extreme Conditions: Leveraging 3D-2D Prior Fusion for Superior Structural and Texture Recovery
Blind Face Restoration under Extreme Conditions: Leveraging 3D-2D Prior Fusion for Superior Structural and Texture Recovery Open
Blind face restoration under extreme conditions involves reconstructing high-quality face images from severely degraded inputs. These input images are often in poor quality and have extreme facial poses, leading to errors in facial structu…
View article: Unfinished debate: Why IPH-based metrics are still needed—An Editorial for “Signal intensity and volume of carotid intraplaque hemorrhage on magnetic resonance imaging and the risk of ipsilateral cerebrovascular events: the Plaque At RISK (PARISK) study”
Unfinished debate: Why IPH-based metrics are still needed—An Editorial for “Signal intensity and volume of carotid intraplaque hemorrhage on magnetic resonance imaging and the risk of ipsilateral cerebrovascular events: the Plaque At RISK (PARISK) study” Open
View article: Extracranial carotid artery atherosclerotic plaque and APOE polymorphisms: a systematic review and meta-analysis
Extracranial carotid artery atherosclerotic plaque and APOE polymorphisms: a systematic review and meta-analysis Open
Introduction Carotid atherosclerotic plaque is an important independent risk factor for stroke. Apolipoprotein E (APOE) influences cholesterol levels and certain isoforms are associated with increased carotid atherosclerosis, though the ex…
View article: Carotid Plaque-RADS
Carotid Plaque-RADS Open
View article: Editorial for “Remodeling in Aortic Stenosis With Reduced and Preserved Ejection Fraction: Insight on Motion Abnormality Via <scp>3D</scp>+Time Personalized <scp>LV</scp> Modelling in Cardiac <scp>MRI</scp>”
Editorial for “Remodeling in Aortic Stenosis With Reduced and Preserved Ejection Fraction: Insight on Motion Abnormality Via <span>3D</span>+Time Personalized <span>LV</span> Modelling in Cardiac <span>MRI</span>” Open
Level of Evidence 5 Technical Efficacy Stage 2
View article: Why is the winner the best?
Why is the winner the best? Open
International benchmarking competitions have become fundamental for the comparative performance assessment of image analysis methods. However, little attention has been given to investigating what can be learnt from these competitions. Do …
View article: Differences in left and right carotid plaque vulnerability in patients with bilateral carotid plaques: a CARE-II study
Differences in left and right carotid plaque vulnerability in patients with bilateral carotid plaques: a CARE-II study Open
Background and purpose Atherosclerosis is a very complex process influenced by various systemic and local factors. Therefore, in patients with bilateral carotid plaques (BCPs), there may be differences in carotid plaque vulnerability betwe…
View article: Morphological and Compositional Features of Chronic Internal Carotid Artery Occlusion in MR Vessel Wall Imaging Predict Successful Endovascular Recanalization
Morphological and Compositional Features of Chronic Internal Carotid Artery Occlusion in MR Vessel Wall Imaging Predict Successful Endovascular Recanalization Open
Background: We sought to determine if the morphological and compositional features of chronic internal carotid artery occlusion (CICAO), as assessed by MR vessel wall imaging (MR-VWI), initially predict successful endovascular recanalizati…