|
|
| |
| Bone Abnormalities Detection and Classification Using Machine Learning Techniques – Literature Review
Download PDF
|
Alaa M. A. Barhoom, Mohammed Rasheed J. AL-Hiealy, and Samy S. Abu-Naser
|
Abstract:
Bone problems is meant to be any of the injuries that affect human bones. Bones injuries are a major cause of abnormalities of the human skeletal system. Although physical injury, causing fracture, controls over disease, nonetheless fracture is one of a few common causes of bone problems. Bone fracture can circuitously cause death, as fractures and their associated difficulties can in some situation produce a descending spiral in health. About 20% of the hip fracture patients passed away within a year of the fracture. The goal of the study was to carry out a literature review of bone abnormalities that used machine learning as a mean of detection and classification of bone abnormalities, by reviewing technique which included study reference, approaches, dataset, programming language used, and the top accuracy achieved. 17 studies published between 2001 and 2020 were designated for analysis. Seven of which employ X-ray images, five employ CT images and five employ MRI Images. The outcomes indicated that amongst the classification methods studied, artificial neural networks, Convolutional Neural Network and Residual neural network extensively used with average accuracy between 73.40% and 94.71%. Most of the designated studies employed X-ray images. The result is that it is possible in the future to develop a deep learning model to identify bone abnormalities with more accurate outcomes, by adjusting the methods used and adding more actual processes.
|
|
|
| | |