Strategic Use of Big Data in Spine Surgery - Testing Digital Prediction of Outcome in Clinical Praxis

This study uses artificial intelligence and big data to predict surgical outcomes for patients with two common spinal conditions: lumbar spinal stenosis (narrowing of the spinal canal in the lower back) and cervical radiculopathy (nerve compression in the neck). Back and neck surgery carries real risks, and outcomes vary widely — predicting in advance which patients will benefit most from surgery could help surgeons and patients make better decisions. Eligible participants are adults aged 18 or older who are referred for surgery for one of these two conditions, have had recent MRI confirming their diagnosis, and provide informed consent. Those whose symptoms turn out to be caused by another condition or who cannot be enrolled within one month of the initial consultation are excluded. Participants complete digital questionnaires before and after surgery, and their outcomes are compared against AI model predictions to test how accurately the tool performs in clinical practice. Spinal surgery decisions are often difficult, with patients and doctors navigating genuine uncertainty. A validated prediction tool could reduce unnecessary surgeries, spare patients from procedures unlikely to help them, and help direct surgical resources toward those who will benefit most — improving both patient outcomes and healthcare efficiency.