Computer-aided diagnosis of lumbar disc herniation

Computer-aided diagnosis systems have been the focus of many research endeavors. They are based on the idea of processing and analyzing various types of input (such as patients medical history, physical examination results, images of different parts of the human body, etc.) to generate a quick and accurate diagnosis. In this work, we propose a system that follows the aforementioned approach to diagnose lumbar disk herniation from a top-down Magnetic Resonance Imaging (MRI) spine view of the suspected region. To the best of our knowledge, this is the first work to consider this type of images for the diagnosis of lumbar disk herniation. The proposed system consists of several stages that include image acquisition and annotation, Region Of Interest (ROI) extraction and enhancement, feature extraction, and classification. The experiments conducted to evaluate the system show that the system is quick and accurate making it a great aid in the diagnosis process as well as an invaluable platform for educational and research purposes.