ANALYSIS OF DENTAL CARIES FROM INTRA-ORAL PERIAPICAL RADIOGRAPHS USING MACHINE LEARNING MODELS

Authors

  • MATHUVANTI K.K Bachelor of dental surgery, Department of Public Health Dentistry, S.R.M. Dental College, Ramapuram, Chennai, India
  • Prabu D Professor and Head of Department, Department of Public Health Dentistry, S.R.M. Dental College, Ramapuram, Chennai, Indiallege
  • SINDHU R Senior lecturer, Department of Public Health Dentistry, S.R.M. Dental College, Ramapuram, Chennai, India
  • DINESH DHAMODHAR Reader, Department of Public Health Dentistry, S.R.M. Dental College, Ramapuram, Chennai, India
  • RAJMOHAN M Reader, Department of Public Health Dentistry, S.R.M. Dental College, Ramapuram, Chennai, India
  • BHARATHWAJ V.V Senior lecturer, Department of Public Health Dentistry, S.R.M. Dental College, Ramapuram, Chennai, India
  • SATHIYAPRIYA S Senior lecturer, Department of Public Health Dentistry, S.R.M. Dental College, Ramapuram, Chennai, India
  • VISHALI M Postgraduate student, Department of Public Health Dentistry, S.R.M. Dental College, Ramapuram, Chennai, India

Keywords:

dental caries, artificial intelligence, machine learning, deep learning technique

Abstract

Background:

With the advancement of technology, dentists can improve their performance in identifying various dental caries. In addition, Artificial Intelligence techniques such as Machine learning methods provide a second opinion for dentists on the task of detecting caries.

A.I.M.:

The study aimed to apply caries prediction through machine learning was analyzed in this paper through an experimental analysis.

Materials And Method:

For this study, there was a data set of 700 X-rays gathered from the different age groups .500 X-rays for training the model and 200 X-rays for testing the model. In addition, seven other machine learning techniques: SVM, non-linearSVM+pca, K.N.N., KNN+pca, Decision tree, Decision tree+depth and two deep learning techniques: Custom CNN, Inception net, were applied to this data set.

Result:

We have compared the performance of our proposed system with certified dentists for marking dental caries with working examples. After analogizing the performance of nine different classifiers, the results show that for carries detection CNN method performs better than other methods.

Conclusion:

To date, for the detection of dental caries, visual and clinical examination and diagnostic aids, which are the IOPA's commonly called the X-rays are in practice. This sequela of procedures helps the dentist to identify the cause and treat accordingly. But it is said that in certain cases, dentists miss the appropriate diagnosis if presented with the bitewing radiographs. So in that cases, when the CNN method of machine learning is used in dentistry, it would be a great help to the dentist in case of appropriate diagnosis, and in normal circumstances also would always help the dentist to recheck the diagnosis whether the diagnosis made is fair and accurate.

Keywords:

Dental caries, artificial intelligence, machine learning, and deep learning technique.

Downloads

Published

2022-08-21

How to Cite

MATHUVANTI K.K, Prabu D, SINDHU R, DINESH DHAMODHAR, RAJMOHAN M, BHARATHWAJ V.V, SATHIYAPRIYA S, & VISHALI M. (2022). ANALYSIS OF DENTAL CARIES FROM INTRA-ORAL PERIAPICAL RADIOGRAPHS USING MACHINE LEARNING MODELS. International Journal of Dental and Clinical Study, 3(3), 01-09. Retrieved from https://ijdcs.com/index.php/ijdcs/article/view/69

Issue

Section

Articles

Most read articles by the same author(s)