Yu Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yu Zhang has a total of 37 co-authors. Here are the top co-authors by number of shared publications:
Xiangwei Luo
Department of Medical Imaging, The 901st Hospital of the Joint Logistics Support Force of PLA, Anhui, China
2 shared publications- Xuefei Deng
School of Basic Medical Sciences, Anhui Medical University, Hefei 230032, China
2 shared publications Bin Hu
Department of Radiology, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China
1 shared publication- Bin Liu
Department of Radiology, The First Affiliated Hospital of Anhui Medical University, Hefei, China
1 shared publication Changsheng Zhou
Department of Radiology, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China
1 shared publicationChen Chen
Department of Radiology, 901st Hospital of the Joint Logistic Support Force of the People’s Liberation Army of China
1 shared publication
Latest Publications
- Open Access
- 3 September 2026
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 13, 1797905
Preliminary exploration of the value of pericoronary adipose tissue radiomics in identifying high-risk patients with chronic coronary syndrome: a retrospective Chinese cohort study
Background: Chronic coronary syndrome (CCS) requires accurate risk stratification for optimized management. Pericoronary adipose tissue (PCAT) inflammation is closely associated with plaque progression, and coronary computed tomography angiography (CCTA)-based radiomics may provide information beyond the traditional fat attenuation index (FAI). Objective: To assess the value of PCAT radiomics and its combination with one-stop CCTA parameters for identifying CCS patients who are classified as high-risk by the 2024 European Society of Cardiology (ESC) guideline-recommended risk factor-clinical (RF-CL) model. Methods: We retrospectively enrolled 175 CCS patients who underwent CCTA between January 2024 and January 2026. According to the 2024 ESC guideline-recommended RF-CL model, patients were classified as high risk (>15%, n = 77) or low risk (≤15%, n = 98). PCAT radiomics features were extracted from the three major coronary arteries, and FAI, computed tomography-derived fractional flow reserve (CT-FFR), coronary artery calcium score (CACS), and the Coronary Artery Disease Reporting and Data System (CAD-RADS) were analyzed. Results: Significant between-group differences were found in right coronary artery (RCA) FAI, left anterior descending (LAD) FAI, CACS, CAD-RADS, distal CT-FFR at the most stenotic plaque, and distal vessel CT-FFR, but not left circumflex (LCX) FAI. Distal vessel CT-FFR showed better discrimination than lesion-specific CT-FFR (area under the receiver operating characteristic curve [AUC] 0.782 vs. 0.698). The RCA PCAT radiomics model outperformed RCA FAI in the testing set (AUC 0.729 vs. 0.620). The integrated model achieved the best performance, with AUCs of 0.860 and 0.823 in the training and testing sets. Conclusion: PCAT radiomics, especially when integrated with one-stop CCTA parameters, improves the identification of CCS patients classified as high-risk by the RF-CL model. Prospective studies with hard cardiovascular outcomes are required to validate its prognostic value and to determine whether this imaging-based strategy can guide therapeutic decision-making.
- 3 September 2026
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 13, 1797905
Open AccessPreliminary exploration of the value of pericoronary adipose tissue radiomics in identifying high-risk patients with chronic coronary syndrome: a retrospective Chinese cohort study
Background: Chronic coronary syndrome (CCS) requires accurate risk stratification for optimized management. Pericoronary adipose tissue (PCAT) inflammation is closely associated with plaque progression, and coronary computed tomography angiography (CCTA)-based radiomics may provide information beyond the traditional fat attenuation index (FAI). Objective: To assess the value of PCAT radiomics and its combination with one-stop CCTA parameters for identifying CCS patients who are classified as high-risk by the 2024 European Society of Cardiology (ESC) guideline-recommended risk factor-clinical (RF-CL) model. Methods: We retrospectively enrolled 175 CCS patients who underwent CCTA between January 2024 and January 2026. According to the 2024 ESC guideline-recommended RF-CL model, patients were classified as high risk (>15%, n = 77) or low risk (≤15%, n = 98). PCAT radiomics features were extracted from the three major coronary arteries, and FAI, computed tomography-derived fractional flow reserve (CT-FFR), coronary artery calcium score (CACS), and the Coronary Artery Disease Reporting and Data System (CAD-RADS) were analyzed. Results: Significant between-group differences were found in right coronary artery (RCA) FAI, left anterior descending (LAD) FAI, CACS, CAD-RADS, distal CT-FFR at the most stenotic plaque, and distal vessel CT-FFR, but not left circumflex (LCX) FAI. Distal vessel CT-FFR showed better discrimination than lesion-specific CT-FFR (area under the receiver operating characteristic curve [AUC] 0.782 vs. 0.698). The RCA PCAT radiomics model outperformed RCA FAI in the testing set (AUC 0.729 vs. 0.620). The integrated model achieved the best performance, with AUCs of 0.860 and 0.823 in the training and testing sets. Conclusion: PCAT radiomics, especially when integrated with one-stop CCTA parameters, improves the identification of CCS patients classified as high-risk by the RF-CL model. Prospective studies with hard cardiovascular outcomes are required to validate its prognostic value and to determine whether this imaging-based strategy can guide therapeutic decision-making.
- 6 February 2026
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
Multi-task Nomogram Model for Predicting Rebleeding and Survival Outcomes in Cerebral Contusion
This study aimed to develop a multi-task nomogram model based on CT imaging to accurately predict rebleeding and survival prognosis in patients with cerebral contusion, thereby enhancing early risk assessment and personalized treatment strategies. A retrospective cohort of 427 patients with CT-confirmed cerebral contusions was analyzed. The study integrated clinical data (e.g., Glasgow Coma Scale scores), radiomics features extracted from CT images, and deep transfer learning (DTL) outputs using a DenseNet121 architecture. A multi-task nomogram model was constructed to concurrently predict rebleeding risk and survival outcomes. Model performance was evaluated using ROC curves, calibration plots, decision curve analysis (DCA), and Harrell's concordance index (C-index). Interpretability was examined using Gradient-weighted Class Activation Mapping (Grad-CAM). The nomogram model exhibited excellent predictive performance, achieving areas under the curve (AUC) values of 0.973 for the training cohort and 0.959 for the testing cohort regarding rebleeding prediction. It also showed the highest prognostic accuracy (C-index 0.857, p < 0.0001). The model highlighted the critical role of GCS scores, particularly in moderate TBI cases (GCS 6-8), where timely intervention improved survival rates by 20%. Grad-CAM visualization confirmed the model's ability to localize hemorrhage regions accurately. A multi-task nomogram model, combining clinical, radiomic, and DTL features, provides a robust tool for early prediction of rebleeding and survival in cerebral contusion patients. Its integration with GCS scores facilitates targeted interventions, especially for moderate cerebral contusion cases, underscoring its clinical utility in improving outcomes.
- 6 February 2026
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
Multi-task Nomogram Model for Predicting Rebleeding and Survival Outcomes in Cerebral Contusion
This study aimed to develop a multi-task nomogram model based on CT imaging to accurately predict rebleeding and survival prognosis in patients with cerebral contusion, thereby enhancing early risk assessment and personalized treatment strategies. A retrospective cohort of 427 patients with CT-confirmed cerebral contusions was analyzed. The study integrated clinical data (e.g., Glasgow Coma Scale scores), radiomics features extracted from CT images, and deep transfer learning (DTL) outputs using a DenseNet121 architecture. A multi-task nomogram model was constructed to concurrently predict rebleeding risk and survival outcomes. Model performance was evaluated using ROC curves, calibration plots, decision curve analysis (DCA), and Harrell's concordance index (C-index). Interpretability was examined using Gradient-weighted Class Activation Mapping (Grad-CAM). The nomogram model exhibited excellent predictive performance, achieving areas under the curve (AUC) values of 0.973 for the training cohort and 0.959 for the testing cohort regarding rebleeding prediction. It also showed the highest prognostic accuracy (C-index 0.857, p < 0.0001). The model highlighted the critical role of GCS scores, particularly in moderate TBI cases (GCS 6-8), where timely intervention improved survival rates by 20%. Grad-CAM visualization confirmed the model's ability to localize hemorrhage regions accurately. A multi-task nomogram model, combining clinical, radiomic, and DTL features, provides a robust tool for early prediction of rebleeding and survival in cerebral contusion patients. Its integration with GCS scores facilitates targeted interventions, especially for moderate cerebral contusion cases, underscoring its clinical utility in improving outcomes.
- Open Access
- 9 April 2025
- Published by Springer Nature in Chinese Journal of Academic Radiology
- Vol. 8 (2) , 156-165
Development and validation of a simple-to-use coronary calcium score based nomograph for predicting obstructive coronary artery disease
- 9 April 2025
- Published by Springer Nature in Chinese Journal of Academic Radiology
- Vol. 8 (2) , 156-165
Open AccessDevelopment and validation of a simple-to-use coronary calcium score based nomograph for predicting obstructive coronary artery disease
- Open Access
- 1 January 2022
- Published by Elsevier in Journal of Clinical Neuroscience
- Vol. 95, 151-158
Parameters of MR perfusion-weighted imaging predict the response and prognosis to high-dose methotrexate-based chemotherapy in immunocompetent patients with primary central nervous system lymphoma
- 1 January 2022
- Published by Elsevier in Journal of Clinical Neuroscience
- Vol. 95, 151-158
Open AccessParameters of MR perfusion-weighted imaging predict the response and prognosis to high-dose methotrexate-based chemotherapy in immunocompetent patients with primary central nervous system lymphoma