Establishing a nomogram to predict refracture after percutaneous kyphoplasty by logistic regression

• DOI: 10.3389/fninf.2023.1304248
• Publication Date: 2023-12-21T04:33:32Z
• Authors: Aiqi Zhang; Hongye Fu; Junjie Wang; Zhe Chen; Jiajun Fan
• Keywords: Nomogram; Lasso; Logistic model tree

Introduction Several studies have examined the risk factors for post-percutaneous kyphoplasty (PKP) refractures and developed many clinical prognostic models. However, no prior research exists using Random Forest (RF) model, a favored tool model development, to predict occurrence of new vertebral compression fractures (NVCFs). Therefore, this study aimed investigate post-PKP fractures, compare predictive performance logistic regression RF models in forecasting visualize model. Methods We collected data from 349 patients who underwent PKP treatment at our institution January 2018 December 2021. Lasso was employed select associated with NVCFs. Subsequently, were established, their capabilities compared. Finally, nomogram created. Results The variables selected regression, including bone density, cement distribution, fracture location, preoperative height, height restoration rate, included both area under curves 0.868 0.786, respectively, training set 0.786 0.599, validation set. Furthermore, calibration curve also outperformed that Conclusion provided better identifying than