Classification of binary fracture using CNN.

Chittajallu, Sai Manvitha; Mandalaneni, Navya Lakshmi Deepthi; Parasa, Dhanush; Bano, Shahana

Abstract

One of the major problems faced by any living organism since infancy are musculoskeletal injuries. To keep it quite simple musculoskeletal injuries are a range of disorders involving muscles, bones, tendons, blood vessels, nerves and other soft tissues. However one of the most common forms of musculoskeletal injuries are fractures. Fractures are one of the most prevalent sores that are faced by any living organism. They are also easily overlooked by the best of physicians. Even with the help of an X-ray, they are one of the hardest symptoms to diagnose. We believe that we can provide a solution to this problem by implementing convolutional neural networks (CNN) image processing algorithms into the field of medicine. We have designed a model using three layers of architecture which has been properly trained to identify the X-ray images that have fractures. To accomplish this we used large datasets that consist of 200 images of human hands, ribs, legs and neck. These large datasets are clearly segregated to identify those images which contain fractures from those images which are perfectly fine. The results gave us accurate predictions using some graphical representations as well as epochs of the various patients.

Citation

CHITTAJALLU, S.M., MANDALANENI, N.L.D., PARASA, D. and BANO, S. 2019. Classification of binary fracture using CNN. In Proceedings of the 1st Global conference for advancement in technology (GCAT 2019), 18-20 October 2019, Bangalore, India. Piscataway: IEEE [online]. Available from: https://doi.org/10.1109/GCAT47503.2019.8978468

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 1st Global conference for advancement in technology (GCAT 2019)
• Start Date: Oct 18, 2019
• End Date: Oct 20, 2019
• Acceptance Date: May 21, 2019
• Online Publication Date: Feb 3, 2020
• Publication Date: Dec 31, 2019
• Deposit Date: Sep 20, 2023
• Publicly Available Date: Sep 20, 2023
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• ISBN: 9781728136950
• DOI: https://doi.org/10.1109/GCAT47503.2019.8978468
• Keywords: Medical imaging; Image recognition; Convolutional neural networks; Fractures; Medical technology; Machine learning
• Public URL: https://rgu-repository.worktribe.com/output/2064095