Puxin Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Puxin Yang has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Wei Chen
Trauma Emergency Center, the Third Hospital of Hebei Medical University, Shijiazhuang, People’s Republic of China
5 shared publications - Baorui Xing
Cangzhou Hospital of integrated traditional Chinese medicine and Western medicine, Cangzhou, China
2 shared publications - Hao Liu
Third Hospital of Hebei Medical University, Shijiazhuang, China
2 shared publications - Tianrui Wang
Department of Orthopedics, The Affiliated Hospital of Qingdao University, Qingdao 266061, PR China
2 shared publications - Yanbin Zhu
Department of Orthopaedic Surgery, Hebei Medical University Third Hospital, Shijiazhuang, China
2 shared publications - Yubin Long
Baoding First Central Hospital, Baoding, China
2 shared publications
Latest Publications
- 23 September 2026
- Published by Springer Nature in International Orthopaedics
- Vol. 50 (12) , 3307-3318
Anticipating femoral shaft nonunion at 1-3 months: a multimodal framework with automated X‑ray segmentation despite metal implants
Background Nonunion is a costly and serious complication in trauma surgery. The RUST score can assess tibial healing on radiographs but offers no prediction of nonunion at nine to 12 months. A prediction model integrating demographics, blood tests, and radiomics fills this gap and brings clinical value. Methods We collected 1-3 month postoperative X-ray, patient data, and blood tests across three centers including 550 patients and 2,344 X-ray. 13 predictive features were selected via two-stage LASSO with group-wise preselection. A U-Net based segmentation framework standardized bone and implant masks under heterogeneous imaging conditions, enabling robust radiomics extraction. ExtraTrees, LightGBM, SVM, CatBoost, Random forest, Gradient boosting, XGBoost, HistGradientBoosting, MLP, Logistic Regression, ElasticNet, Lasso, AdaBoost, Gaussian NB, and Decision Tree were employed to develop NU model. Results A total of 492 patients formed the training cohort and 58 patients formed the external validation cohort. An automated segmentation model achieved mean Dice coefficients of 0.947 for bone/callus and 0.921 for implants. LASSO selected 13 predictors: five radiomic features, seven blood tests, and one clinical variable. The ExtraTrees-based NU model yielded an AUC of 0.891±0.021 in the training cohort, with good calibration and net benefit on decision curve analysis. In external validation, AUC was 0.840. The model enables early nonunion risk stratification within three months after surgery. Conclusion The NU model is the first automated multimodal prediction model integrating radiomics, blood tests, and clinical data to predict femoral shaft nonunion within three months after intramedullary nailing, providing a technically scalable solution for early nonunion risk stratification before conventional imaging becomes diagnostic.
- 23 September 2026
- Published by Springer Nature in International Orthopaedics
- Vol. 50 (12) , 3307-3318
Anticipating femoral shaft nonunion at 1-3 months: a multimodal framework with automated X‑ray segmentation despite metal implants
Background Nonunion is a costly and serious complication in trauma surgery. The RUST score can assess tibial healing on radiographs but offers no prediction of nonunion at nine to 12 months. A prediction model integrating demographics, blood tests, and radiomics fills this gap and brings clinical value. Methods We collected 1-3 month postoperative X-ray, patient data, and blood tests across three centers including 550 patients and 2,344 X-ray. 13 predictive features were selected via two-stage LASSO with group-wise preselection. A U-Net based segmentation framework standardized bone and implant masks under heterogeneous imaging conditions, enabling robust radiomics extraction. ExtraTrees, LightGBM, SVM, CatBoost, Random forest, Gradient boosting, XGBoost, HistGradientBoosting, MLP, Logistic Regression, ElasticNet, Lasso, AdaBoost, Gaussian NB, and Decision Tree were employed to develop NU model. Results A total of 492 patients formed the training cohort and 58 patients formed the external validation cohort. An automated segmentation model achieved mean Dice coefficients of 0.947 for bone/callus and 0.921 for implants. LASSO selected 13 predictors: five radiomic features, seven blood tests, and one clinical variable. The ExtraTrees-based NU model yielded an AUC of 0.891±0.021 in the training cohort, with good calibration and net benefit on decision curve analysis. In external validation, AUC was 0.840. The model enables early nonunion risk stratification within three months after surgery. Conclusion The NU model is the first automated multimodal prediction model integrating radiomics, blood tests, and clinical data to predict femoral shaft nonunion within three months after intramedullary nailing, providing a technically scalable solution for early nonunion risk stratification before conventional imaging becomes diagnostic.
- Open Access
- 31 July 2026
- Published by Springer Nature in BMC Musculoskeletal Disorders
- Vol. 27 (1)
Post-injury high-sensitivity C-reactive protein to lymphocyte ratio and nonunion risk in adult extremity fractures: evidence from a multicenter retrospective cohort
- 31 July 2026
- Published by Springer Nature in BMC Musculoskeletal Disorders
- Vol. 27 (1)
Open AccessPost-injury high-sensitivity C-reactive protein to lymphocyte ratio and nonunion risk in adult extremity fractures: evidence from a multicenter retrospective cohort
- Open Access
- 22 August 2025
- Published by Wolters Kluwer Health in International Journal of Surgery
- Vol. 111 (12) , 8943-8952
Association between post-trauma platelet-lymphocyte ratio and nonunion in patients with extremity fractures: a multicenter retrospective cohort study
Background: Post-traumatic coagulation and inflammatory activation are associated with various clinical outcomes, and impairments of hematoma formation and inflammatory cell recruitment in the early phases potentially adversely affect bone healing. We aimed to explore and quantify the association between serum platelet-lymphocyte ratio (PLR) levels and postoperative nonunion in patients undergoing extremity fracture Surgery. Methods: This retrospective, multicenter cohort study analyzed data from five tertiary hospitals, including adult patients who underwent surgery for limb fractures between January 2020 and December 2022. The exposure was the post-trauma platelet-to-lymphocyte ratio (PLR), calculated by dividing the platelet count by the lymphocyte count. The outcome of interest was fracture nonunion, defined as the failure of fracture healing for a minimum of 9 months, without evidence of progression toward healing over the preceding 3 months, which was confirmed through a comprehensive review of medical records, telephone follow-ups, and outpatient assessments. Dose–response analysis using restricted cubic spline (RCS) function and multivariate logistic regression were conducted to evaluate the relationship between PLR and nonunion risk. Results: In total, 30 791 patients were included, with an average age of 49.1 ± 16.5 years, and 60.5% were males. Nonunion occurred in 656 patients (2.1%). RCS analysis showed a significant negative non-linear dose–response relationship between PLR and nonunion risk ( P -nonlinearity < 0.001, P -overall < 0.001), and the cutoff value for the risk-protective effect was 145.0. Compared to high PLR levels (≥145.0), lower PLR levels (P < 0.001); and this risk effect remained evident after adjustment (adjusted odds ratio = 1.265; 95% CI, 1.057–1.515; P = 0.010). Results are robust across several prespecified sensitivity or exploratory analyses. Subgroup analyses revealed heterogeneity of association according to surgery site and fixation methods ( P values for interaction < 0.050). Conclusion: This study demonstrates that reduced post-trauma PLR levels are associated with an increased risk of nonunion, serving as a potential biomarker for risk stratification and personalized intervention planning during the critical bone healing window.
- 22 August 2025
- Published by Wolters Kluwer Health in International Journal of Surgery
- Vol. 111 (12) , 8943-8952
Open AccessAssociation between post-trauma platelet-lymphocyte ratio and nonunion in patients with extremity fractures: a multicenter retrospective cohort study
Background: Post-traumatic coagulation and inflammatory activation are associated with various clinical outcomes, and impairments of hematoma formation and inflammatory cell recruitment in the early phases potentially adversely affect bone healing. We aimed to explore and quantify the association between serum platelet-lymphocyte ratio (PLR) levels and postoperative nonunion in patients undergoing extremity fracture Surgery. Methods: This retrospective, multicenter cohort study analyzed data from five tertiary hospitals, including adult patients who underwent surgery for limb fractures between January 2020 and December 2022. The exposure was the post-trauma platelet-to-lymphocyte ratio (PLR), calculated by dividing the platelet count by the lymphocyte count. The outcome of interest was fracture nonunion, defined as the failure of fracture healing for a minimum of 9 months, without evidence of progression toward healing over the preceding 3 months, which was confirmed through a comprehensive review of medical records, telephone follow-ups, and outpatient assessments. Dose–response analysis using restricted cubic spline (RCS) function and multivariate logistic regression were conducted to evaluate the relationship between PLR and nonunion risk. Results: In total, 30 791 patients were included, with an average age of 49.1 ± 16.5 years, and 60.5% were males. Nonunion occurred in 656 patients (2.1%). RCS analysis showed a significant negative non-linear dose–response relationship between PLR and nonunion risk ( P -nonlinearity < 0.001, P -overall < 0.001), and the cutoff value for the risk-protective effect was 145.0. Compared to high PLR levels (≥145.0), lower PLR levels (P < 0.001); and this risk effect remained evident after adjustment (adjusted odds ratio = 1.265; 95% CI, 1.057–1.515; P = 0.010). Results are robust across several prespecified sensitivity or exploratory analyses. Subgroup analyses revealed heterogeneity of association according to surgery site and fixation methods ( P values for interaction < 0.050). Conclusion: This study demonstrates that reduced post-trauma PLR levels are associated with an increased risk of nonunion, serving as a potential biomarker for risk stratification and personalized intervention planning during the critical bone healing window.