XWXiaochun Wang

Xiaochun Wang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xiaochun Wang has a total of 95 co-authors. Here are the top co-authors by number of shared publications:

  • Yan Li

    Department of Radiology, The First Hospital of Shanxi Medical University, Taiyuan, China

    12 shared publications
  • Bin Wang

    Department of Radiology, The First Hospital of Shanxi Medical University, Taiyuan, China

    10 shared publications
  • Yongfang Wang

    Department of Radiology, The First Hospital of Shanxi Medical University, Taiyuan, China

    10 shared publications
  • Zhichang Fan

    Department of Radiology, The First Hospital of Shanxi Medical University, Taiyuan, China

    8 shared publications
  • Guoqiang Yang

    Department of Radiology, First Hospital of Shanxi Medical University

    6 shared publications
  • Junting Guo

    Department of Medical Imaging, Shanxi Medical University, Taiyuan, China

    5 shared publications

Latest Publications

  • Open Access

    Pathology-guided multimodal transformer predicts postoperative survival benefit from adjuvant chemotherapy in glioma: from DNA damage repair to immune characterization

    by Zehui Li, Xin Duan, Junjie Lian, Xuan Li, Qian Liang, Xu Liu, Yan Tan, Guoqiang Yang, Jiangfeng Du, Lina Wu, et al.

    Introduction: High-grade gliomas (HGGs) are aggressive primary brain tumors with high recurrence rates and poor prognosis. Survival benefit from adjuvant temozolomide (TMZ) varies substantially across patients, and reliable preoperative biomarkers for treatment stratification remain limited. This study aims to develop an interpretable multimodal deep learning model integrating whole-slide histopathological images and multiparametric MRI to preoperatively predict survival benefit from adjuvant TMZ in HGG patients. Methods: A Transformer-based multimodal fusion model was proposed. Key tumor regions were identified from whole-slide pathology images via weakly supervised clustering, and pathological features were extracted using attention-based multiple instance learning. Concurrently, 2.5D tumor subregional features were derived from multiparametric MRI sequences. These multimodal features were deeply fused using a Transformer encoder. The model was developed and validated on a multicenter cohort (504 patients), and its biological basis was explored through transcriptomic analysis, gene set enrichment analysis, protein-protein interaction networks, and immune deconvolution. Results: The fusion model achieved AUCs of 0.910, 0.861, and 0.833 in the training, internal validation, and external validation cohorts, respectively. Transcriptomic analysis implicated DNA damage repair processes in high-benefit patients, suggesting that enhanced DNA repair capacity may contribute to TMZ sensitivity. Immune deconvolution showed higher M1 macrophage infiltration and lower regulatory T cell abundance in the high-benefit group, consistent with a less immunosuppressive tumor microenvironment. Importantly, low-benefit patients who completed TMZ chemotherapy had shorter survival than those treated with concurrent chemoradiotherapy alone or surgery alone. Conclusion: This model identifies patients unlikely to benefit from standard TMZ chemotherapy, providing a tool for postoperative personalized treatment stratification and optimization of therapeutic strategies in HGGs.

  • Application of transformer-enhanced convolutional neural network: multicenter MRI assessment of muscle invasion in bladder cancer

    by Zhichang Fan, Ding Li, Wenjing Chen, Yan Li, Junting Guo, Wenqiao Zheng, Bin Wang, Yongfang Wang, Xiaochun Wang

    Objective Accurate preoperative assessment of muscle invasion in bladder cancer (BCa) guides therapy selection. However, MRI interpretation varies across readers and lesion morphologies. Therefore, we aimed to overcome the morphology-associated diagnostic bias through a deep learning method. Materials and methods This multicenter study included 1374 patients with BCa. An nnU-Net was fine-tuned to assist in lesion segmentation on T2-weighted images, providing inputs for a 2.5D ConvNeXt-tiny model to assess muscle invasion. The performance of the model was compared between pedunculated and sessile lesions. Furthermore, a head-to-head comparison was conducted among the model, a senior radiologist, and a junior radiologist. Results The validation Dice coefficient of nnU-net was 0.834. In the validation and three prospective test sets, the ConvNeXt-tiny model achieved areas under the receiver-operating characteristic curve of 0.915–0.925 for identifying muscle invasion in BCa, with accuracies of 84.9–91.0%, sensitivities of 81.3–96.2%, and specificities of 81.1–93.8%. In the subgroup analysis of pedunculated and sessile lesions, the model’s diagnostic performance showed no significant difference across all datasets. In contrast, the two radiologists’ specificities declined from around 90% in pedunculated lesions to approximately 75% (p = 0.010–0.050) in sessile lesions across the validation set, internal test set, and external test set 1. Therefore, in the head-to-head comparison of sessile lesions, the model demonstrated significantly higher specificities (91.9–96.0%) than the two radiologists (72.8–79.8%, p < 0.001–0.005) across the same datasets. Conclusions The DL model assesses muscle invasion in BCa independently of lesion morphology and holds potential for clinical application, particularly in sessile lesions. Key Points Question Conventional MRI assessment of muscle invasion risk in pedunculated and sessile bladder cancers may be biased, but clear evidence and potential solutions are still lacking. Findings Morphology-associated diagnostic bias indeed exists, mainly as overstaging of sessile bladder cancer, while deep learning assessment of muscle invasion risk is morphology-independent. Clinical relevance The transformer-enhanced convolutional neural network developed in this study effectively reduces overestimation of muscle invasion risk in sessile bladder cancer and may serve as a complementary imaging tool for clinical evaluation. Graphical Abstract

  • Intracranial Plaque Characteristics and Extracranial Carotid Plaque‐ RADS Associated With Stroke Recurrence: A High‐Resolution Vessel Wall Imaging Study

    by Wenqiao Zheng, Yichun Qu, Fanyu Du, Zhichang Fan, Yan Li, Bin Wang, Yongfang Wang, Le Wang, Xiaochun Wang

    Background: In addition to intracranial plaques, extracranial carotid plaques have also been linked to stroke recurrence. However, the association of combined intracranial and extracranial plaque characteristics with stroke recurrence remains unclear. Purpose: To evaluate the association between intracranial plaque features and extracranial carotid Plaque Reporting and Data System (Plaque‐RADS) with stroke recurrence utilizing high‐resolution vessel wall imaging (HR‐VWI). Study Type: Retrospective. Population: The 388 intracranial atherosclerotic ischemic stroke patients (mean age, 57.3 ± 11.5 years, 274 males). Field Strength/Sequence: 3T, three‐dimensional T1 weighted 3D fast spin echo. Assessment: HR‐VWI was performed in all patients within 7 days of the stroke onset. The imaging features assessed included intracranial plaque characteristics (degree of stenosis, plaque burden, enhancement ratio, remodeling index, and intraplaque hemorrhage [IPH]) and extracranial carotid Plaque‐RADS. Patients enrolled between May 2022 and July 2024 were included in the study. All patients were followed for a minimum of 12 months. Statistical Tests: Mann–Whitney U test or χ 2 test, univariate and multivariate Cox regression analyses, time‐dependent ROC and AUC curves, and Kaplan–Meier survival curves. p ‐values less than 0.05 were regarded as statistically significant. Results: During a median follow‐up period of 24 months, 55 patients experienced recurrent stroke. Intracranial plaque enhancement ratio (HR, 1.62; 95% CI, 1.11–2.38), IPH (HR, 2.55; 95% CI, 1.47–4.40), and extracranial carotid Plaque‐RADS (HR, 2.23; 95% CI, 1.29–3.86) were significantly associated with stroke recurrence. Time‐dependent ROC indicated that the maximum AUCs for enhancement ratio, IPH, and Plaque‐RADS were 0.75, 0.69, and 0.68, respectively, while the Cox model including all three reached a maximum AUC of 0.79. Data Conclusion: The combination of intracranial plaque characteristics and extracranial carotid Plaque‐RADS can be used to assess ischemic stroke recurrence. Evidence Level: 3. Technical Efficacy: Stage 2.


  • Radiogenomics of muscle-invasive bladder cancer: machine learning for predicting prognostically relevant genetic mutations

    by Hua Wei, Zhichang Fan, Yan Li, Wenqiao Zheng, Xiaoyue Zhang, Zeke Chen, Wenxin Li, Jiangfeng Du, Xiaochun Wang

  • Preoperative prediction of parametrial invasion in early-stage cervical cancer: a radiomics nomogram fusing multi-parametric MRI and clinical biomarkers

    by Bin Wang, Jianxin Xiao, Yuanyuan Ding, Jinjie Yu, Lizhi Xie, Aiguo Zhou, Canyu Wang, Xiaochun Wang, Yongfang Wang

    Purpose This study aimed to develop and validate a radiomics nomogram that integrates multi-parametric MRI and clinical factors for the preoperative prediction of parametrial invasion (PMI) in early-stage cervical cancer (ECC). Materials and methods A total of 363 patients with ECC (FIGO stages IB–IIA) were divided into training, internal validation, and external validation cohorts. All patients underwent T2WI, DWI, and T1c scans before radical hysterectomy. Radiomics features were extracted from T2WI, DWI, and T1c images, and selected using the max-relevance and min-redundancy (mRMR) method and the least absolute shrinkage and selection operator (LASSO). Radiomics signatures were then derived from these selected features. An MRI model was built using the radiomics signatures to evaluate their performance in distinguishing patients with PMI. A radiomics nomogram was constructed based on the optimal radiomics signature, pre-procedure hematocrit levels, and CA-125 levels. The discrimination performance of the nomogram was subsequently evaluated. Results For the MRI model, the radiomics signatures yielded AUCs of 0.834 (95% CI: 0.7275–0.9399) and 0.800 (95% CI: 0.6902–0.9105) in the internal and external validation cohorts, respectively. The radiomics nomogram, which integrated the radiomics signatures from T2WI, DWI, and T1c, along with hematocrit and CA-125 levels, showed excellent discrimination between PMI and non-PMI groups. The nomogram achieved an AUC of 0.827 (95% CI: 0.7116–0.9430) in the internal validation cohort and 0.806 (95% CI: 0.6997–0.9114) in the external validation cohort. The specificity and sensitivity were 0.866 and 0.762, respectively, in the internal validation cohort, and 0.875 and 0.583 in the external validation cohort. Conclusions The developed radiomics nomogram provides a non-invasive and reliable tool for preoperative PMI prediction in ECC. By facilitating more accurate risk stratification, it has the potential to inform personalized therapeutic planning.


  • Open Access

    Combined evidence from artificial neural networks and human brain-lesion models reveals that language modulates vision in human perception

    by Haoyang Chen, Bo Liu, Shuyue Wang, Xiaosha Wang, Wenjuan Han, Xiaochun Wang, Yixin Zhu, Yanchao Bi

    Comparing information structures in between deep neural networks (DNNs) and the human brain has become a key method for exploring their similarities and differences. Recent research has shown better alignment of vision–language DNN models, such as contrastive language–image pretraining (CLIP), with the activity of the human ventral occipitotemporal cortex (VOTC) than earlier vision models, supporting the idea that language modulates human visual perception. However, interpreting the results from such comparisons is inherently limited owing to the ‘black box’ nature of DNNs. Here we combine model–brain fitness analyses with human brain lesion data to examine how disrupting the communication pathway between the visual and language systems causally affects the ability of vision–language DNNs to explain the activity of the VOTC to address this. Across four diverse datasets, CLIP consistently captured unique variance in VOTC neural representations, relative to both label-supervised (ResNet) and unsupervised (MoCo) models. This advantage tended to be left-lateralized at the group level, aligning with the human language network. Analyses of 33 patients who experienced a stroke revealed that reduced white matter integrity between the VOTC and the language region in the left angular gyrus was correlated with decreased CLIP–brain correspondence and increased MoCo–brain correspondence, indicating a dynamic influence of language processing on the activity of the VOTC. These findings support the integration of language modulation in neurocognitive models of human vision, reinforcing concepts from vision–language DNN models. The sensitivity of model–brain similarity to specific brain lesions demonstrates that leveraging the manipulation of the human brain is a promising framework for evaluating and developing brain-like computer models.


  • Open Access

    DWI and ADC Habitat Imaging in Predicting HER2 Expression Status in Bladder Cancer: A Retrospective Study

    by Zeke Chen, Zhichang Fan, Xiaoyue Zhang, Wenxin Li, Yan Li, Bin Wang, Yongfang Wang, Guoqiang Yang, Xiaochun Wang

    Background: Human epidermal growth factor receptor 2 (HER2) antibody-coupled drugs have shown promising clinical benefits in patients with bladder cancer (BCa). HER2 expression status is generally detected clinically using postoperative pathological immunohistochemistry (IHC), but preoperative non-invasive detection of BCa HER2 expression status remains to be sought. The aim of this study was to investigate the value of diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) habitat imaging in predicting the expression of HER2 in BCa. Methods: This retrospective study included 232 BCa patients (November 2022–February 2024) with HER2 status confirmed by immunohistochemistry. The K-means clustering algorithm is used to re-segment the region of interest. Based on the spatial distribution of the habitat map, the histogram features of each subregion were extracted. Based on the Spearman correlation coefficient (> 0.7) feature screening results, a support vector machine (SVM) classification model was established to predict the expression of HER2 in BCa. The discrimination ability of the model was evaluated by plotting the receiver operating characteristic (ROC) curve and calculating the area under the curve (AUC), and the diagnostic performance of the model was comprehensively evaluated by combining the calibration curve and the decision curve. Results: Randomly divided patients into training cohort (N = 148, median age 68.66 years; 121men), validation (N = 47, median age 69.12 years; 39 men), and test cohort (N = 37, median age 67.92 years; 32men) according to the ratio of 6:2:2. Based on the contour coefficient, K = 2 is finally selected as the clustering parameter to cluster the DWI and ADC images into two subregions. A total of 80 features were extracted from the four sub-regions of the two sequences. After screening, an SVM prediction model was constructed from the remaining 17 features. In the SVM model, the AUC of the training set was 0.88 (95% CI: 0.82–0.93), the validation set was 0.85 (95% CI: 0.72–0.94), and the test set was 0.84 (95% CI: 0.88–0.94). Conclusion: MRI-based habitat analysis can help distinguish heterogeneous regions of BCa and effectively predict HER2 expression status of BCa. Clinical trial number: Not applicable.

  • Open Access

    Parameter-driven habitat imaging based on intravoxel incoherent motion MRI for preoperative prediction of muscle invasion in bladder cancer

    by Junting Guo, Zhichang Fan, Ding Li, Qianmeng Chen, Wenlong Qin, Yan Li, Bin Wang, Yongfang Wang, Xiaochun Wang

    To develop a parameter-driven habitat imaging based on intravoxel incoherent motion (IVIM) MRI and to validate its effectiveness in preoperatively predicting muscle invasion in bladder cancer (BC). This prospective study enrolled 693 pathology-confirmed BC patients who preoperatively underwent multi-b-value diffusion-weighted imaging (DWI). DWI images were acquired using a 3.0 T MRI scanner with 11 b values. Patients were randomly divided into training and testing sets (7:3). K-means clustering was applied to diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) maps to assign all voxels to distinct intra-tumoral habitats. The voxel proportion (p) and corresponding IVIM values were extracted for each habitat. Features were selected using intraclass correlation coefficient and Pearson correlation analysis. Three random forest models were then constructed: a habitat model, a clinical-radiologic model, and a combined model integrating both. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), and Shapley additive explanations (SHAP) was used to interpret feature contributions. Two habitats (K = 2) were identified, including a low perfusion and restricted diffusion (LP-RD) habitat with decreased D, D* and f, and a high perfusion and unrestricted diffusion (HP-UD) habitat with increased D, D* and f. The habitat model achieved AUCs of 0.950 and 0.814 in the training and testing sets, respectively. Incorporating clinical-radiologic factors into the combined model improved its performance, reaching AUC values of 0.987 and 0.832. SHAP analysis identified the p and D* values in the HP-UD habitat, and tumor location as the top contributors to the combined model’s predictions. The parameter-driven habitat imaging based on IVIM MRI offers a promising and noninvasive tool for preoperative assessment of muscle invasion in BC. It may help guide treatment stratification and reduce unnecessary interventions. Future studies will validate these findings in multi-center cohorts and advance integration into clinical workflows.