Preliminary analysis shows transfer learning improves detection accuracy of pulp calcifications in dental radiographs, indicating potential in endodontic treatment.
Key Points
To evaluate transfer learning models for detecting pulp calcifications in dental radiographs.
Collected 240 cropped panoramic radiographs, 120 with pulp calcifications and 120 without.
Preprocessed images using CLAHE and data augmentation techniques.
Utilized VGG16, ResNet101V2, and MobileNetV2 pre-trained models for image classification.
VGG16 achieved training, validation, and test accuracies of 0.80, 0.85, and 0.85, respectively.
VGG16 showed a precision of 0.84, a recall of 0.87, and an F1-score of 0.86.