Comparative Analysis of Deep Learning Architectures for Bone Fracture Classification

Abstract

The examination of bone fractures found in Xray images by radiologists is both time-consuming and prone to human error. The use of deep learning models to accurately and quickly diagnose bone fractures can assist doctors in treating patients. In this study, various state-of-the-art deep learning models were applied to the FracAtlas dataset utilizing effective data augmentation techniques to classify bone fractures with high accuracy. The performance of the deep learning models used was compared, and the proposed approach achieved an accuracy of ${9 2 . 4 7 \%}$, demonstrating competitive results with existing studies.

Ähnliche Arbeiten