Analysis of Dental radiograph using visual computing techniques
Machine Learning approaches are increasingly prevalent in the field of medical image engineering. We are seeing the development of ever improving image-based diagnostics, disease prognostics, and risk assessment techniques. This group has identified various areas with potential of development
2025-06-28 16:25:06 - Adil Khan
Analysis of Dental radiograph using visual computing techniques
Project Area of Specialization Artificial IntelligenceProject SummaryMachine Learning approaches are increasingly prevalent in the field of medical image engineering. We are seeing the development of ever improving image-based diagnostics, disease prognostics, and risk assessment techniques. This group has identified various areas with potential of development of medical imaging techniques in the area of dental radiographs after literature review. This group intends to develop a machine learning-based system to detect teeth in a dental OPG X-ray, and identify various artefacts like cavities and root-canals etc. We would be working in collaboration with a leading medical institute based in Islamabad, to acquire a large dataset of dental radiographs that will be pre-processed, catalogued, and annotated before they will be used to develop and train the ML-based platform. We believe that this work has the potential for immensely impactful research contributions. While a lot of work is being done in analyzing OPGs using visual computing techniques, there is still a lot of room for improvement in terms of accuracy and latencies. This research wants to contribute to the development of framework for the analysis of dental radiographs for analysis and possibly diagnosis.
Project ObjectivesThe aim of the project is to test and tune various Machine Learning algorithms and also includes the analysis of dental radiographs that are used in the clinical diagnosis, treatment, and surgery of patients. Hidden dental structures, malignant or benign tumors, bone loss, and cavities can all be discovered with radiography. Dental radiography analysis is required for diagnostic and treatment operations such as root canal therapy, caries diagnosis, and orthodontic diagnosis and treatment planning. This researcher proposes to develop a platform that would use the state-of-the-art in machine learning tools to analyses dental OPG radiographs and use them to identify and diagnose any anomalies and diseases that can normally be detected by a trained dental specialist. The aim is to develop a framework that would allow the development of a low-cost computer/smart-phone application that would be able to diagnose dental ailments and then suggest treatment options to help out the dentist. If utilized, such a system can help mitigate the problem of misdiagnosis, allow a low-cost second opinion, and may even encourage patients from the low-income bracket to pursue dental treatment once a serious ailment is detected.
We collected 220 OPG’s from dental institute that includes the dental radiographs of males and females (children, adults, and aged). Most of the radiographs are of adults, around 60%. 30% are of aged/old and 10% of children. After collection of the dental radiographs data, we stared the process of Semantic Segmentation on a MATLAB. We increased the number of images using data augmentation technique. We separated the 220 images into training and testing images.
? A deep understanding of image processing especially in the context of application to neural networks
? Development of a pre-processing protocol customized for dental panoramic radiographs that aims to facilitate the optimal extraction of each tooth for further analysis
? Contribute to on-going work at VC-Tech in the classification of individual teeth and diagnosis of one or more pathologies
? Publish our findings in a peer-reviewed journal to contribute to the field of visual computing at large.
? A deep understanding of image processing especially in the context of application to neural networks
? Development of a pre-processing protocol customized for dental panoramic radiographs that aims to facilitate the optimal extraction of each tooth for further analysis
? Contribute to on-going work at VC-Tech in the classification of individual teeth and diagnosis of one or more pathologies
? Publish our findings in a peer-reviewed journal to contribute to the field of visual computing at large.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 70000 | |||
| . | Equipment | 2 | 35000 | 70000 |