Artificial Intelligence in Predicting Surgical Problems and Postoperative Morbidity in Mandibular Third Molar Extractions.
Authors
Affiliations (8)
Affiliations (8)
- Department of Dentistry, JGMMMC, Hubballi, Karnataka, India.
- Oral and Maxillofacial Surgeon, King Abdullah Medical City, Makkah, Saudi Arabia.
- Department of Oral and Maxillofacial Surgery, Faculty of Dental Sciences, Uttar Pradesh University of Medical Sciences, Etawah, Uttar Pradesh, India.
- Department of Public Health, College of Applied Medical Sciences, Qassim University, Buraydah, Saudi Arabia.
- Department of Periodontics, Saveetha Institute of Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, Tamil Nadu, India.
- Department of Oral and Maxillofacial Surgery, Qassim University, Buraydah, Saudi Arabia.
- Department of Oral Medicine and Radiology, Kalinga Institute of Dental Sciences, KIIT Deemed to be University, Bhubaneswar, Odisha, India.
- Department of Oral Maxillofacial Surgery and Diagnostic Sciences, College of Dentistry, King Faisal University, Al Hofuf, Saudi Arabia.
Abstract
Mandibular third molar (MTM) extractions are among the most frequent oral surgical procedures, often associated with variable surgical difficulty and postoperative complications. Artificial intelligence (AI) offers potential in predicting these outcomes to enhance preoperative planning. This study was done to evaluate the utility of AI in predicting surgical difficulty and postoperative morbidity in MTM extractions using cone-beam computed tomography (CBCT) data and patient variables. A cross-sectional study of 40 patients undergoing MTM extraction was conducted. AI algorithms (random forest and convolutional neural networks) analyzed CBCT features and clinical parameters to predict difficulty and morbidity. Statistical evaluation was done using SPSS version 26. The AI model demonstrated an overall accuracy of 87.5% in predicting surgical difficulty and 82.3% for postoperative morbidity. Significant predictors included root morphology, proximity to the inferior alveolar nerve, and patient age (P < 0.05). AI-based models can effectively predict MTM extraction difficulty and postoperative morbidity, facilitating personalized treatment planning.