Kruger, N. and Abramowitz, S. and Nitschke, G. (2022) Machine Learning in Diagnosing Cervical Spine Injuries, Global Spine Journal, 12, 4S-204S, Sage journals.
Text
2022-Machine_20Learning_20in_20Diagnosing_20Cervical_20Spine_20Injuries.pdf Download (494kB) |
Abstract
Machine-learning algorithms (Artificial Intel ligence) have demonstrated remarkable progress in image recognition tasks, especially in the medical field. In our set ting, radiologist reporting on x-rays is often not available in peripheral hospitals. X-rays often need to be interpreted by junior doctors working after hours in busy emergency departments, leaving room for radiological errors. AI could prove to be the ideal diagnostic tool where swift and ac curate diagnosis of cervical spine injuries are required. Machine-learning networks originally developed for other tasks can be applied to skeletal x-rays with minimal in tervention. Machine-learning is increasingly being used in diagnosis and can be expected to gradually change clinical practice, assisting clinicians, and improving inter-rater reliability. We aimed to evaluate the diagnostic accuracy of AI in interpreting lateral cervical spine x-rays.
Item Type: | Journal article (paginated) |
---|---|
Subjects: | Computing methodologies > Artificial intelligence |
Date Deposited: | 02 Nov 2023 15:44 |
Last Modified: | 02 Nov 2023 15:44 |
URI: | https://pubs.cs.uct.ac.za/id/eprint/1587 |
Actions (login required)
View Item |