Diagnostic Prediction Model for Encephalitis and Intracranial Malignant Tumors Based on Common Blood Test Indicators and Random Forest Algorithm

Model parameters were optimized using 10-fold cross-validation and grid search. Model performance was evaluated using accuracy, precision, recall, F1-score, and the area under the ROC curve (AUC). Feature importance was analyzed using SHAP values.Results The LASSO regression ultimately identified 14 key features.

The constructed random forest model demonstrated good predictive performance on the test set, with an AUC of 0.9512, accuracy of 0.9080, recall of 0.8261, and F1 score of 0.9048. SHAP analysis revealed that C‑reactive protein (CRP) contributed the most to the model, followed by lymphocyte (LYM) and phosphorus (P).Conclusion The random forest model based on common blood indicators can effectively differentiate encephalitis from malignant tumors, exhibiting high clinical practical value, with CRP being a particularly important predictor.