IADR Abstract Archives
APPLICATION of MACHINE LEARNING on PERIAPICAL DISEASE DIAGNOSIS on X-RAY IMAGES
Objectives: We conducted this research to determine sensitivity, specificity and accuracy of machine learning in peri-apical
diseases diagnosis on X- Ray images.
Methods: In our research, the model is constructed by applying Faster R-CNN combining with experts’ knowledge. Experts have important roles in collecting data, labelling the lesion regions and choosing images for testing.
- Step 1: Parameter selecting
- Step 2: Model training
- Step 3: Model testing
- Step 4: Model evaluating
Results: Endodontist diagnosed 143 teeth with periapical lesions and 377 teeth without periapical lesions. For software’s diagnosis, the sensitivity, specificity and accuracy is 89.5%, 97.9% and 95.6 %, respectively.
Conclusions: As can be seen from this study, artificial intelligent with faster R-CNN training can predict apical lesions properly with favorable result.
diseases diagnosis on X- Ray images.
Methods: In our research, the model is constructed by applying Faster R-CNN combining with experts’ knowledge. Experts have important roles in collecting data, labelling the lesion regions and choosing images for testing.
- Step 1: Parameter selecting
- Step 2: Model training
- Step 3: Model testing
- Step 4: Model evaluating
Results: Endodontist diagnosed 143 teeth with periapical lesions and 377 teeth without periapical lesions. For software’s diagnosis, the sensitivity, specificity and accuracy is 89.5%, 97.9% and 95.6 %, respectively.
Conclusions: As can be seen from this study, artificial intelligent with faster R-CNN training can predict apical lesions properly with favorable result.