Lumbar Spinal Stenosis Analysis with Deep Learning Based Decision Support Systems

Authors : Sinan ALTUN, Ahmet ALKAN

Pages : 1200-1215

Doi:10.35378/gujs.1116423

View : 170 | Download : 509

Publication Date : 2023-09-01

Article Type : Research Paper

Abstract :Lumbar spinal stenosis insert ignore into journalissuearticles values(LSS); is a condition that affects the quality of life of the 3 vertebrae, the disc and the canal in the lower back. In this region, the nerves in the canal may be subjected to pressure for various reasons, and disease occurs. Surgical intervention is required to treat canal stenosis, and the exact location and size of the spinal stenosis is critical to the surgery. The UNet model, which is an example of this network, can be further deepened with various deep learning networks. In this study, it will be the basis for creating a system that helps in the diagnosis of spinal stenosis by using a deeper network. The ResUNET model using ResNet as the backbone achieved an average IoU of 0.987. This study demonstrated that expert decision support systems using MR images can be used in the diagnosis of LSS.
Keywords : Unet, Deep learning, Image processing, Lumbar spinal stenosis, Semantic segmentation