Abstract:

Purpose: The purpose of this study is to improve and evaluate the function of diagnostic computer software designed to assess dental deficiencies in panoramic radiography. The data set consists of 153 images, including 99 full teeth and 54 missing teeth for the detection of missing teeth. All images have been re-checkled and verified by oral, dental and chen radiology specialists. All images in the data set were re-dimensional to 971 x 474 pixels before training. A open source python programming language and a random sequence has been created using OpenCV, NumPy, Pandas, and Matplotlib libraries effectively. A previously trained Google Net Inception v3 CNN network was used for preliminary processing and data sets were trained using transfer learning. Results: The success rate for the model estimation of the images used in education is 94.7%. The estimated success rate of images assigned to the test not used in training is 75%. The result: As the data set increases in deep learning techniques, the success rates also increase. The success rates will increase in the training models of the data set that will be formed with more images. The future work should be done with larger data sets. Keywords Panoramic radiography, deep learning, artificial intelligence

Keywords: