Aim: The aim of this study was to construct a risk prediction model for intraoperative acquired pressure injury in neurosurgical patients and to verify the predictive effect of the model, with the intention of providing a reference for the prevention of IAPI in neurosurgical patients. Methods: The clinical data of 776 children who underwent neurosurgery in a tertiary grade A children\'s specialized hospital in Chongqing from January to June 2023 were retrospectively collected. Variables were screened using LASSO regression, and nine machine learning algorithms were selected. The machine learning model was constructed and validated using ten-fold cross-validation and a validation set. The predictive effects of each model were compared using ROC curves, cross-validation results, and efficacy indicators. The optimal model for predicting IAPI in neurosurgical patients was constructed. Results: The incidence of IAPI in neurosurgical patients was 7 .9%. The XGBoost algorithm modelhad the best predictive performance, with sensitivity , specificity , accuracy of 0.8889, 0.4813, 0.126, respectively . The clinical decision curve showed that the predicted probability of IAPI occurrence by this model was consistent with the actual occurrence probability . It was found that clinical indicators such as surgical blood loss, surgical position, patient age, operation duration, anesthesia duration, and the use of external force were identified as key risk factors. Conclusion: The neurosurgical patient prediction model constructed in this study has certain predictive efficacy and can provide a scientific basis for clinical nurses to identify high-risk neurosurgical patients with IAPI and take personalized preventive measures in an early stage.