Automated Dental Image Analysis by Deep Learning on Small Dataset


Dental radiography gives vital confirmation to clinical determination, treatment and quality assessment. Much exertion has been spent on creating digitalized dental X-beam picture investigation frameworks for clinical quality change. In this paper, we present the datasets, systems, and results directed to assess dental treatment characteristics utilizing periapical dental X-beam pictures taken when the tasks. So as to help dental practitioners to settle on clinical choices, we propose an instrument pipeline for mechanized clinical quality assessment.

We manufacture a dataset with 196 patients’ periapical dental radiography pictures previously, then after the fact the medicines. Radiography pictures are named as cases that are ‘showing signs of improvement’, ‘deteriorating’ and ‘have no express change’ by assigned dental specialists. Our proposition incorporates a programmed technique with the restorative information to edit the ROIs for clinical assessment – the apical adjoining districts, and after that sets of ROIs are nourished into a CNN to prepare the model for mechanized clinical quality assessment. Our methodology accomplishes the F1 score of 0.749, which is equivalent to the execution of master dental practitioners and radiologist.

BASE PAPER: Automated Dental Image Analysis by Deep Learning on Small Dataset

Leave a Reply

Your email address will not be published. Required fields are marked *