Machine Learning Model for Predicting Postoperative Pain in Cases of Irreversible Pulpitis

Area of Science:

• Dentistry
• Machine Learning
• Pain Management

Background:

• Postoperative pain is a common complication following endodontic therapy.
• Predicting and managing this pain is crucial for patient well-being and treatment success.

Purpose of the Study:

• To develop and validate supervised machine learning models for predicting postoperative pain after endodontic treatment in irreversible pulpitis cases.
• To identify key clinical predictors influencing pain occurrence.

Main Methods:

• A prospective cohort of 354 patients undergoing endodontic treatment was analyzed.
• Eight supervised machine learning algorithms were trained and validated using a 70/30 train-test split.
• Performance was assessed using Area Under the Curve (AUC), accuracy, precision, recall, and F1-score, with class imbalance addressed via Synthetic Minority Oversampling Technique.

Main Results:

• Supervised machine learning models demonstrated good predictive performance for postoperative pain.
• Logistic Regression achieved the best performance at 24 hours (AUC 0.74), while Support Vector Machine excelled at 72 hours (AUC 0.81).
• Patient age and sex were identified as the most significant predictors of pain.

Conclusions:

• Supervised machine learning offers a promising approach for predicting postoperative endodontic pain.
• Logistic Regression and Support Vector Machine models show potential for clinical application in personalized pain management.
• These models can aid clinicians in anticipating and managing patient pain more effectively.