Motoki Fukuda1, Yoshiko Ariji2, Yoshitaka Kise2, Michihito Nozawa2, Chiaki Kuwada2, Takuma Funakoshi2, Chisako Muramatsu3, Hiroshi Fujita4, Akitoshi Katsumata5, Eiichiro Ariji2. 1. Department of Oral and Maxillofacial Radiology, Aichi-Gakuin University School of Dentistry, Nagoya, Japan. Electronic address: halpop@dpc.agu.ac.jp. 2. Department of Oral and Maxillofacial Radiology, Aichi-Gakuin University School of Dentistry, Nagoya, Japan. 3. Faculty of Data Science, Shiga University, Shiga, Japan. 4. Department of Electrical, Electronic and Computer Faculty of Engineering, Gifu University, Gifu, Japan. 5. Department of Oral Radiology, Asahi University School of Dentistry, Mizuho, Japan.
Abstract
OBJECTIVE: The aim of this study was to compare time and storage space requirements, diagnostic performance, and consistency among 3 image recognition convolutional neural networks (CNNs) in the evaluation of the relationships between the mandibular third molar and the mandibular canal on panoramic radiographs. STUDY DESIGN: Of 600 panoramic radiographs, 300 each were assigned to noncontact and contact groups based on the relationship between the mandibular third molar and the mandibular canal. The CNNs were trained twice by using cropped image patches with sizes of 70 × 70 pixels and 140 × 140 pixels. Time and storage space were measured for each system. Accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC) were determined. Intra-CNN and inter-CNN consistency values were calculated. RESULTS: Time and storage space requirements depended on the depth of CNN layers and number of learned parameters, respectively. The highest AUC values ranged from 0.88 to 0.93 in the CNNs created by 70 × 70 pixel patches, but there were no significant differences in diagnostic performance among any of the models with smaller patches. Intra-CNN and inter-CNN consistency values were good or very good for all CNNs. CONCLUSIONS: The size of the image patches should be carefully determined to ensure acquisition of high diagnostic performance and consistency.
OBJECTIVE: The aim of this study was to compare time and storage space requirements, diagnostic performance, and consistency among 3 image recognition convolutional neural networks (CNNs) in the evaluation of the relationships between the mandibular third molar and the mandibular canal on panoramic radiographs. STUDY DESIGN: Of 600 panoramic radiographs, 300 each were assigned to noncontact and contact groups based on the relationship between the mandibular third molar and the mandibular canal. The CNNs were trained twice by using cropped image patches with sizes of 70 × 70 pixels and 140 × 140 pixels. Time and storage space were measured for each system. Accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC) were determined. Intra-CNN and inter-CNN consistency values were calculated. RESULTS: Time and storage space requirements depended on the depth of CNN layers and number of learned parameters, respectively. The highest AUC values ranged from 0.88 to 0.93 in the CNNs created by 70 × 70 pixel patches, but there were no significant differences in diagnostic performance among any of the models with smaller patches. Intra-CNN and inter-CNN consistency values were good or very good for all CNNs. CONCLUSIONS: The size of the image patches should be carefully determined to ensure acquisition of high diagnostic performance and consistency.