Blog
A deep learning model based on concatenation approach to predict the time to extract a mandibular third molar tooth
This study proposes a practical deep learning model to predict the time required for mandibular third molar extraction by combining panoramic radiographic images and clinical data. A concatenation approach integrating a Convolutional Neural Network (CNN) and a Multilayer Perceptron (MLP) was developed to simulate clinical decision-making, in which both radiographic appearance and patient-related factors are considered. Using 724 panoramic radiographs and associated clinical variables, the combined model demonstrated strong correlation with actual extraction times, outperforming models based on imaging or clinical data alone. The results suggest that multimodal AI models may support surgical planning, especially for less experienced clinicians.
Predicting the time required for mandibular third molar extraction is a critical step in preoperative planning in oral and maxillofacial surgery. Extraction time is closely associated with surgical difficulty, operator stress, patient discomfort, and the risk of intra- and postoperative complications. In daily clinical practice, experienced surgeons intuitively estimate extraction time by combining radiographic findings with patient-specific and operator-related factors, whereas less experienced clinicians often struggle with this assessment. This study demonstrates that a multimodal deep learning model, integrating panoramic radiographs with clinical variables, can approximate this clinical reasoning process and provide an objective estimation of extraction time. Such a tool may support decision-making in scheduling, case selection, referral to specialists, and patient counseling, particularly for novice dentists and residents. By offering a transparent and explainable prediction framework, this approach has the potential to enhance surgical planning, optimize resource allocation, and improve patient safety without replacing clinical judgment.