Performance of deep learning object detection technology in the detection and diagnosis of maxillary sinus lesions on panoramic radiographs.

Objective: The first aim of this study was to determine the performance of a deep learning object detection technique in the detection of maxillary sinuses on panoramic radiographs. The second aim was to clarify the performance in the classification of maxillary sinus lesions compared with healthy maxillary sinuses.

Methods: The imaging data for healthy maxillary sinuses (587 sinuses, Class 0), inflamed maxillary sinuses (416 sinuses, Class 1), cysts of maxillary sinus regions (171 sinuses, Class 2) were assigned to training, testing 1, and testing 2 data sets.

Quantification of voxel values in micro computed tomography using multiple porosity hydroxyapatite blocks.

Microfocus X-ray computed tomography (micro-CT) has been applied as a method for the nondestructive and detailed assessment of trabecular bone patterns and tooth structure. Voxel values obtained from micro-CT are not absolute values. Therefore, voxel values were assessed using hydroxyapatite (HA) blocks with a different vesicle rate to quantify voxel values of micro-CT images in the present investigation.