Evaluating the performance of generative adversarial network-synthesized periapical images in classifying C-shaped root canals

Abstract

This study evaluated the performance of generative adversarial network (GAN)-synthesized periapical images for classifying C-shaped root canals, which are challenging to diagnose because of their complex morphology. GANs have emerged as a promising technique for generating realistic images, offering a potential solution for data augmentation in scenarios with limited training datasets. Periapical images were synthesized using the StyleGAN2-ADA framework, and their quality was evaluated based on the average Frechet inception distance (FID) and the visual Turing test. The average FID was found to be 35.353 (± 4.386) for synthesized C-shaped canal images and 25.471 (± 2.779) for non C-shaped canal images. The visual Turing test conducted by two radiologists on 100 randomly selected images revealed that distinguishing between real and synthetic images was difficult. These results indicate that GAN-synthesized images exhibit satisfactory visual quality. The classification performance of the neural network, when augmented with GAN data, showed improvements compared with using real data alone, and could be advantageous in addressing data conditions with class imbalance. GAN-generated images have proven to be an effective data augmentation method, addressing the limitations of limited training data and computational resources in diagnosing dental anomalies.

Evaluation of synthetic Raman spectra for use in virus detection

“…Generative Adversarial Networks (GANs) are a type of adversarial deep learning model that can be used to synthesize new data from existing data [7]; though images are the most common application, a GAN may be applied to produce 1-dimensional data such as spectra. GANs were first introduced in 2014 as a method for synthesizing new data by using two models working together [8]: a Generator, which uses the distribution of the real data to produce fake data, and a Discriminator, which aims to differentiate between the real and fake data [9]. The two networks are adversarial in nature, providing feedback to one another to improve each model.…”