Deep Transfer Learning and Majority Voting Approaches for Osteoporosis Classification

Ashames, M.M.Ceylan, MuratJennane, R.2022-05-232022-05-2320212147-6799https://doi.org/10.18201/IJISAE.2021473646https://hdl.handle.net/20.500.13091/2378Osteoporosis is a systemic skeletal disease characterized by low bone mass density and deterioration of the micro-architectural structure of the bone tissue, increasing bone fragility, and the probability of fracture. In this study, we propose a non-invasive method for osteoporosis classification using X-ray images (plain radiographs) of the ankle. Convolutional Neural Networks along with Data Augmentation techniques and Deep Transfer Learning Architectures are combined to classify X-ray images of healthy and osteoporotic patients. The proposed approach achieved an accuracy of 99% using ResNet50, and 100% with GoogleNet. © 2021, Ismail Saritas. All rights reserved.eninfo:eu-repo/semantics/openAccessCNNData augmentationOsteoporosisTransfer learningX-rayDeep Transfer Learning and Majority Voting Approaches for Osteoporosis ClassificationArticle10.18201/IJISAE.20214736462-s2.0-85124485434