Multiclass classification of infections after cervical spine surgery in the elderly: a machine learning approach based on preoperative and perioperative data

Key Points

• Infections following cervical spine surgery can be accurately classified using a machine learning approach.
• Using preoperative and perioperative data, the model achieved a notable accuracy rate of 85% in predicting infection types.
• Observational analysis utilizing a multiclass classification algorithm assessed various preoperative and perioperative factors.
• This study highlights the potential for machine learning to improve infection diagnosis but requires validation in diverse settings.