Tools
Tools
Daiya, Kajol Deepak (2022) Instance Segmentation for Detecting Dental Caries in Panoramic X-rays using Detectron2. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (2MB) | Preview |
Preview |
PDF (Configuration manual)
Download (1MB) | Preview |
Abstract
Dental disease is referred to as ”silent disease” since it does not cause pain until it has progressed to an advanced level. Dental disease is mostly avoided, but if not detected early, it can develop into periodontal infections and pus. For detecting oral illnesses, dentists rely solely on visual assessment using radiological images. Unfortunately, these radiographs have several drawbacks, including poor image quality, a low diagnosis rate, and a long processing time. This study sought to assist dentists by performing instance segmentation on panoramic Xrays for detecting five classes of caries. Faster RCNN R101-FPN and Faster RCNN X101-FPN pre-trained models are implemented and evaluated to examine the accuracy of the proposed dental caries detection model. The model’s average precision @IOU is 53.512 for segmentation and 66.18 for bbox, this determines how well the detecting bounding boxes match the ground truth bounding boxes. For the development of the dental care industry, the proposed system implemented cutting-edge computing algorithms and compared their results. The results of the experiments reveal that the Detectron2 model has proven to be accurate at recognizing five classes of dental caries on panoramic Xrays.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science R Medicine > RK Dentistry T Technology > Biomedical engineering |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Tamara Malone |
| Date Deposited: | 24 Jan 2023 10:48 |
| Last Modified: | 24 Jan 2023 10:48 |
| URI: | https://norma.ncirl.ie/id/eprint/6105 |
Actions (login required)
 |
View Item |