Automated Analysis of Tooth Anatomy and Pathological Conditions from Orthopantomogram using Deep Neural Networks

Abstract

This research project aims to automate the identification, labeling, and counting of teeth, as well as the classification of abnormalities and detection of caries in dental X-rays, specifically orthopantomograms (OPGs). It involves several deep neural networks and learning algorithms. The first module uses semantic segmentation with a U-net model to create masks for tooth detection, which are then refined with the YOLOv3 detector, achieving 80% accuracy. Canonical correlation analysis (CCA) helps find tooth midpoints and count the total number of teeth. The second module classifies abnormalities and pathologies using transfer learning with the Inceptionv3 model, yielding moderate accuracy. Caries detection is performed with thresholding and segmentation. The third module detects three treated pathologies—root canal treatments, crowns, and implants—using Faster RCNN and Inceptionv3, showing fair accuracy. Overall, the automated approach demonstrates promising results for enhancing X-ray image interpretation and diagnosing oral diseases.

Keywords:

• Histogram equalization
• Object detection
• Orthopantamograms
• Semantic segmentation
• Transfer learning

Authors Contributions

TSA carried out the conceptualization and initial study of the manuscript, and helped in preparing the original draft and editing. VJ contributed to the conceptualization, writing the technical part of the manuscript, and editing. GMKK helped to prepare the original draft, helped in writing the manuscript, and verified the clinical part of the manuscript. PAA helped in verifying the clinical part of the review. All authors helped to bring the final version of the manuscript.

Data availability statement

The data used in this paper are available from the corresponding author upon reasonable request.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Funding

No funding was received for conducting this study. No funds, grants, or other support were received for this manuscript.

Additional information

Notes on contributors

Tuhina Sheryl Abraham

Tuhina Sheryl Abraham received her BE degree in bio-medical engineering from Jerusalem College of Engineering (Affiliated to Anna University), Chennai, India in 2020 and ME degree in medical electronics from Sri Sivasubramaniya Nadar College of Engineering (Affiliated to Anna University), Kalavakkam, Chennai, India in 2022. Her areas of interest in research are medical image processing, machine learning and deep learning concepts and its applications (AI) in healthcare, AR and VR in medicine and design & innovation. Email: [email protected]

Vijay Jeyakumar

Vijay Jeyakumar is an associate professor in the Department of Biomedical Engineering, Sri Sivasubramaniya Nadar College of Engineering, Chennai, and has more than 14 years of experience in teaching and research. He completed his bachelor's degree in electronics and communication engineering from Anna University in 2005 and obtained his master's degree in medical electronics from the College of Engineering, Guindy Campus in 2008. He did his research in medical image retrieval and obtained his PhD degree in medical informatics from Anna University in 2015. He has published more than 50 papers in peer-reviewed journals and conferences. To his credit, he has authored 12 book chapters in medical image retrieval and analysis, BCI, and Deep learning domains. He is a recognized supervisor of Anna University, Chennai, and a doctoral committee member for several scholars of deemed-to-be Universities. Currently, he is guiding 08 PhD scholars. He has obtained funding from CSIR, TANSCST, and ISTE to organize workshops, seminars, and STTP. Presently, he has been involved in 04 projects worth Rs. 2.48 crores funded by DST, AICTE NASF, and SSN Trust. He is a senior member of IEEE and an active member in societies like ISTE, IE, and IAMI. His five patenting applications are published by Intellectual property India. Corresponding author. Email: [email protected]

Gurucharan Marthi Krishna Kumar

Gurucharan Marthi Krishna Kumar received his BE degree in biomedical engineering from SSN College of Engineering, Anna University, Chennai, India, in 2021. He is currently pursuing a PhD in neuroscience at the Montreal Neurological Institute, McGill University, Montreal, Canada. His areas of interest include medical image processing, diffusion MRI and tractography, deep learning and computer vision applications in healthcare. Email: [email protected]

Ponsekar Abraham Anandapandian

Ponsekar Abraham Anandapandian completed his undergraduate and postgraduate from TN Dr MGR Medical University. Started academic career in the year 2003, when he joined SRM University as senior lecturer. Since then he has mentored undergraduates and post graduates. He has been an executive committee member of the prestigious Indian prosthodontic society for three consecutive years and of the Tamil Nadu & Puducherry state branch of the Indian Prosthodontic Society. He is actively involved in the accreditation process of SRM, Bharath University, Dr MGR Educational & Research institute, Adhiparasakthi Dental college as coordinator for NAAC, ISO & NABH. Currently holds the post of joint director, ACSAMRI, Dean, Academic Staff College, which includes the Medical, Dental and allied intuitions of Dr MGR Educational & research Institute Chennai. In addition, he is also the head the department of Prosthodontics and crown & Bridge in ThaiMoogambigai Dental College, constituent of the University. In addition to the academic commitments he is also involved in active clinical practice for the past 25 years. Email: [email protected]