Rui Zhang

Mr. Rui Zhang

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 January 2024 in IEEE Signal Processing Letters
Verified
Journal Article
Closed access
XYXin Ye
Xin Ye
RJRongxin Jiang
Rongxin Jiang
XTXiang Tian
Xiang Tian
RZRui Zhang
Rui Zhang
YCYaowu Chen
Yaowu Chen
This letter proposes a novel method for effectively utilizing multiple teachers in feature-based knowledge distillation. Our method involves a multi-teacher feature ensemble module for generating a robust feature ensemble and a student-teacher mapping module for bridging the student feature and ensemble feature. In addition, we utilize separate optimization, where the student's feature extractor is optimized under distillation supervision while its classifier is obtained through classifier reconstruction. We evaluate our method on the CIFAR-100, ImageNet and MS-COCO datasets, and the experimental results demonstrate its effectiveness.

2 November 2024 in Bioengineering
Verified
Journal Article
Open access
PDF file
RZRui Zhang
Rui Zhang
MLMiao Lu
Miao Lu
JZJiayuan Zhang
Jiayuan Zhang
XCXiaoyan Chen
Xiaoyan Chen
FZFudong Zhu
Fudong Zhu
YCYuqi Cao
Yuqi Cao
Given the complexity of oral mucosal disease diagnosis and the limitations in the precision of traditional object detection methods, this study aims to develop a high-accuracy artificial intelligence-assisted diagnostic approach based on the SegFormer semantic segmentation model. This method is designed to automatically segment lesion areas in white-light images of oral mucosal diseases, providing objective and quantifiable evidence for clinical diagnosis. This study utilized a dataset of oral mucosal diseases provided by the Affiliated Stomatological Hospital of Zhejiang University School of Medicine, comprising 838 high-resolution images of three diseases: oral lichen planus, oral leukoplakia, and oral submucous fibrosis. These images were annotated at the pixel level by oral specialists using Labelme software (v5.5.0) to construct a semantic segmentation dataset. This study designed a SegFormer model based on the Transformer architecture, employed cross-validation to divide training and testing sets, and compared SegFormer models of different capacities with classical segmentation models such as UNet and DeepLabV3. Quantitative metrics including the Dice coefficient and mIoU were evaluated, and a qualitative visual analysis of the segmentation results was performed to comprehensively assess model performance. The SegFormer-B2 model achieved optimal performance on the test set, with a Dice coefficient of 0.710 and mIoU of 0.786, significantly outperforming other comparative algorithms. The visual results demonstrate that this model could accurately segment the lesion areas of three common oral mucosal diseases. The SegFormer model proposed in this study effectively achieves the precise automatic segmentation of three common oral mucosal diseases, providing a reliable auxiliary tool for clinical diagnosis. It shows promising prospects in improving the efficiency and accuracy of oral mucosal disease diagnosis and has potential clinical application value.

10 June 2026 in BMC Oral Health
Verified
Journal Article
Open access
YCYuqi Cao
Yuqi Cao
MLMiao Lu
Miao Lu
JZJiayuan Zhang
Jiayuan Zhang
RZRui Zhang
Rui Zhang
LSLinjun Shi
Linjun Shi
QCQianming Chen
Qianming Chen
Supervised training for oral potentially malignant disorder (OPMD) image segmentation requires expensive annotated data, particularly scarce in remote regions with few medical specialists. This study aims to propose a self-guided pre-training method to automatically extract features from unlabeled images, enhancing the performance of OPMD lesion segmentation. This study utilized 3,417 OPMD photographs from ZJUSS as the internal dataset and two independent external datasets from WCHS and CS-SJTU for validation. The internal labeled dataset was evaluated using five independent patient-level train/validation/test splits to prevent patient-level data leakage. We proposed a multiscale separable attention masked image model, MS-SAMIM, and a self-guided mask generation method. To enhance robustness, we incorporated a teacher-student consistency constraint and a contrastive learning constraint. Our model achieved 84.35% Dice, 74.65% IoU, 81.84% sensitivity, and 88.02% precision in OPMD lesion segmentation, surpassing various advanced fully supervised and other self-supervised approaches. In the external validation, the model achieved 82.15% Dice, 70.09% IoU, 77.96% sensitivity, and 87.84% precision on external dataset #1 and 81.18% Dice, 69.71% IoU, 78.58% sensitivity, and 84.39% precision on external dataset #2. The proposed self-guided pre-training strategy may reduce reliance on pixel-level annotations and improve OPMD lesion segmentation from clinical photographs. Further prospective validation is required before clinical deployment.

8 June 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
WCWei Chen
Wei Chen
PHPeng Huang
Peng Huang
ZCZi-Jian Cheng
Zi-Jian Cheng
YCYaowu Chen
Yaowu Chen
XTXiang Tian
Xiang Tian
RZRui Zhang
Rui Zhang
(1) Background: Network medicine has reshaped how systemic comorbidities are quantified, but the internal comorbidity structure of oral diseases remains undescribed at fine ICD-10 granularity. (2) Methods: We analyzed 2,863,671 outpatient visit records from 583,614 patients (2011–2025). Using ICD-10 four-character codes (75 disease nodes), comorbidity networks were built for five age strata via relative risk and Bonferroni-corrected Fisher’s exact tests; longitudinal sequences were mined for progression trajectories; and quasi-causal analyses (Cox regression, negative outcome controls, Baron–Kenny mediation) evaluated pathway directionality and specificity. (3) Results: The all-age network contained 75 nodes and 167 edges (modularity = 0.53) forming eight communities. Complexity peaked at 18–29 years and declined with age. Dental caries became the strongest hub in the 60+ group (degree = 9). Adjusted Cox regression confirmed directionality (pulpitis → tooth defect HR = 2.65; caries → pulpitis HR = 2.25), and negative outcome controls confirmed biological specificity. Mediation analysis showed pulpitis completely mediated the caries → tooth defect association (proportion mediated ≈ 100%, 95% CI 90%–128%). An oral mucosal immune cluster (burning mouth syndrome, lichen planus, candidiasis, xerostomia) emerged as a clinically actionable community. (4) Conclusions: Oral diseases form biologically coherent, age-evolving comorbidity communities, with pulpitis as the critical mediating intervention point in the caries-to-tooth-defect cascade, providing a reusable network-medicine substrate for age- and sex-specific risk-stratified disease management.

29 June 2026 in Bioengineering
Verified
Journal Article
Open access
PDF file
WCWei Chen
Wei Chen
PHPeng Huang
Peng Huang
ZCZi-Jian Cheng
Zi-Jian Cheng
YCYaowu Chen
Yaowu Chen
XTXiang Tian
Xiang Tian
RZRui Zhang
Rui Zhang
Background: Network medicine has reshaped how systemic comorbidities are quantified, but the internal comorbidity structure of oral diseases remains undescribed at four-character ICD-10 granularity. Methods: A total of 2,863,671 outpatient visit records from 583,614 patients (2011–2025) were analyzed. Using ICD-10 four-character codes (75 disease nodes), comorbidity networks were constructed for five age strata, with edges selected by relative risk (RR) > 1.5 and Bonferroni-corrected Fisher’s exact tests. Patient-level longitudinal sequences were mined for progression trajectories, and quasi-causal analyses—Cox regression, negative outcome controls, and Baron–Kenny mediation—were used to evaluate pathway directionality and specificity. Results: The all-age network contained 75 nodes and 167 edges (modularity = 0.53), forming eight communities. Network complexity peaked at 18–29 years and declined with age. Dental caries emerged as the strongest hub in the 60+ stratum (degree = 9). Cox regression adjusted for age, sex, and healthcare utilization confirmed pathway directionality (pulpitis → tooth defect: hazard ratio (HR) = 2.65; caries → pulpitis: HR = 2.25), and negative outcome controls confirmed biological specificity. Mediation analysis showed that pulpitis completely mediated the caries → tooth defect association (proportion mediated ≈ 100%; 95% confidence interval (CI), 90–128%). An oral mucosal immune cluster (burning mouth syndrome, lichen planus, candidiasis, and xerostomia) emerged as a clinically actionable community. Conclusions: Oral diseases form biologically coherent, age-evolving comorbidity communities, and pulpitis is the critical mediating intervention point in the caries-to-tooth-defect cascade. The framework provides a reusable network-medicine substrate for age- and sex-specific risk-stratified oral disease management.