An Enhanced Efficientnet Algorithm For Prediction Of Cervical Spine Fracture

Over 1.5 million spine fractures result in 17,730 spinal cord injuries yearly. The cervical spine is where fractures in the spine most frequently occur. Usually, fractures in the cervical spine are caused by car accidents, slips and falls, and conditions like osteoporosis. The majority of the fractures are caused by osteoporosis. Every year, 2 million broken bones in people of all ages occur due to this condition. A cervical spine fracture can cause paralysis, brain damage, and possibly death for the sufferer. A delay in a fracture diagnosis can have terrible repercussions, including worsening of neurologic function and even death. It is crucial to find vertebral fractures quickly. Computed tomography, multidetector computed tomography, magnetic resonance imaging, and other techniques can be used to diagnose spine injuries. The results of these Computed Topography and Multidetector Computed Topography Scans are more precise. We use a state-of-the-art technique to build a model using different Deep neural network algorithms, such as EfficinetNet, Vision Transformer, VGG, Inspection, MobileNet, and Densenet, which will be applied on CT and MDCT scans of patients with spinal cord fractures for foretelling and diagnosing the trauma, paralysis, or catastrophic decline in neurologic function with serious repercussions. The accuracy of the proposed EfficientNet model in evaluating the probability of a cervical spine fracture is 79%.