The goal of this observational study is to establish a clinical prediction model based on artificial intelligence algorithms for patients with postoperative symptomatic spinal epidural hematoma(PSSEH) after spinal surgery. The investigators will collect the features of 2000 patients as the training set to train the models using AI algorithms, then recruit a prospective cohort of 2000 patients to test if the trained model is precise and consistent to predict the occurrence of PSSEH as well as to find out whether there is any new risk factor of PSSEH.

Firstly, the investigators plan to collect the features of 2000 patients who received spinal surgery at Peking University Third Hospital from January 2014 to December 2024, including those with PSSEH and the controls, as the training set to build up a prediction model based on multiple AI algorithms. Then, the investigators will choose the optimal model through comparing various features, assessing the robustness of the model quantitatively using k-fold cross validation and exploring the influence of all features on the output of the final model by means such as SHAP. Lastly, the investigators will recruit a prospective cohort of approximately 2000 patients who receive spinal surgery from January 2025 to December 2026 at Peking University Third Hospital and collect the same features as the training set to test if the trained prediction model is precise and consistent to predict the occurrence of PSSEH and whether there is any feature , undiscovered before, that is strongly related to PSSEH.

Inclusion Criteria:

• patients who will receive spinal surgery from January 2025 to December 2026 at Peking University Third Hospital

Exclusion Criteria:

• patients with a history of lumbar trauma before admission;
• patients who suffer from malignant tumor;
• patients with systemic infection;
• patients who quit the study midway

patients with spinal disorders who come to Peking University Third Hospital for treatment