Transfer Learning-Based Classifier to Automate the Extraction of False X-Ray Images From Hospital's Database

Abstract

Background: During preclinical training, dental students take radiographs of acrylic (plastic) blocks containing extracted patient teeth. With the digitisation of medical records, a central archiving system was created to store and retrieve all x-ray images, regardless of whether they were images of teeth on acrylic blocks, or those from patients. In the early stage of the digitisation process, and due to the immaturity of the data management system, numerous images were mixed up and stored in random locations within a unified archiving system, including patient record files. Filtering out and expunging the undesired training images is imperative as manual searching for such images is problematic. Hence the aim of this stidy was to differentiate intraoral images from artificial images on acrylic blocks.

Methods: An artificial intelligence (AI) solution to automatically differentiate between intraoral radiographs taken of patients and those taken of acrylic blocks was utilised in this study. The concept of transfer learning was applied to a dataset provided by a Dental Hospital.

Results: An accuracy score, F1 score, and a recall score of 98.8%, 99.2%, and 100%, respectively, were achieved using a VGG16 pre-trained model. These results were more sensitive compared to those obtained initally using a baseline model with 96.5%, 97.5%, and 98.9% accuracy score, F1 score, and a recall score respectively.

Conclusions: The proposed system using transfer learning was able to accurately identify "fake" radiographs images and distinguish them from the real intraoral images.

Keywords: Dental radiographs images; Image processing; Transfer learning; VGG16.

Fig. 1

Diagram showing the separation of the merged images datasets.

Fig. 2

Dataset summary.

Fig. 3

Model flowchart diagram.

Fig. 4

Image preprocessing transformations.

Fig. 5

Visual representation of the transfer learning model based on VGG-16.

Fig. 6

(A) Average accuracy per epoch for 10 folds for the baseline model. (B) Average loss per epoch for 10 folds for the baseline model. (C) Average accuracy per epoch for 10 folds of the transfer learning model. (D) Average accuracy per epoch for 10 folds for the baseline model.

Fig. 7

(A) Confusion matrix of best-performing baseline model results on the test dataset. (B) Confusion matrix of best-performing transfer learning model on the test dataset.