Review of Swarm Fuzzy Classifier and a Convolutional Neural Network with VGG-16 Pre-Trained Model on Dental Panoramic Radiograph for Osteoporosis Classification
Author(s): Usman Bello Abubakar, Moussa Mahamat Boukar and Senol Dane*
Abstract
Background: Low bone mineral density and micro-architectural degradation of bone tissue describe osteoporosis, as a bone disease, which increases the risk of fracture. Osteoporosis can be identified, amongst other modalities such as Dual-Energy X-ray Absorptiometry (DXA), by looking at 2D x-ray images of the bone. Through visual clue analysis on trabecular bone structure, dental panoramic radiography (DPR) images provide a relatively affordable tool for evaluating bone density change. To improve diagnostic process and avoid misdiagnosis of medical images, Artificial Intelligence (AI) models especially Convolutional Neural Network (CNN) are employed to manipulate and interpret visual medical data. Objective: The paramount goal of this paper is to provide a performance review and classification accuracy comparison of swarm fuzzy classifier approach and a convolutional neural network with VGG-16 pre-trained model approach on dental panoramic radiograph for osteoporosis classification. Results: The experimental results showed that using CNN with transfer learning pre-trained achieved the accuracy of 84%. The results gotten from the swarm fuzzy classifier indicated a performance classification result to be 98%, for the diagnosis of a low BMD or osteoporosis.