Related Experiment Video
Updated: Aug 30, 2025
05:52
Novel Mini-open Transforaminal Lumbar Interbody Fusion
Published on: June 6, 2025
163
Construct and Validate a Predictive Model for Surgical Site Infection after Posterior Lumbar Interbody Fusion Based
Chuang Xiong1, Runhan Zhao1, Jingtao Xu1
1Department of Orthopedics, The First Affiliated Hospital of Chongqing Medical University, Chongqing 400016, China.
Computational and Mathematical Methods in Medicine
|September 2, 2022
Summary
Machine learning accurately predicts surgical site infections after lumbar fusion. This tool identifies key risk factors like low albumin, diabetes, and dural tears, improving patient care.
Area of Science:
- Neurosurgery
- Medical Informatics
Background:
- Surgical site infection (SSI) is a significant complication following lumbar fusion surgery.
- Early detection and intervention are crucial for mitigating patient harm.
Purpose of the Study:
- To develop and validate a machine learning model for predicting SSIs after posterior lumbar interbody fusion (PLIF).
- To identify critical risk factors associated with SSIs.
- To evaluate the impact of synthetic minority oversampling technique (SMOTE) on model performance.
Main Methods:
- Retrospective review of 584 patients undergoing PLIF for degenerative lumbar disease.
- Data collected included clinical and laboratory information.
- Seven machine learning algorithms were employed, with SMOTE applied to the training set.
- Model performance was validated using a separate validation set.
Main Results:
- The incidence of SSI was 5.65% (33 patients).
- Significant predictors for SSI included low preoperative albumin, diabetes, intraoperative dural tear, and rheumatic disease.
- The AdaBoost Classification Trees model demonstrated superior performance.
- SMOTE enhanced the performance across all tested models.
Conclusions:
- A machine learning model incorporating SMOTE accurately predicts SSIs after PLIF.
- The model aids in clinical decision-making and optimizing perioperative management.
