Clinical machine-learning model for laminoplasty vs. laminectomy in IDEM tumors

Abstract Intradural extramedullary (IDEM) spinal tumors are commonly managed with laminectomy or laminoplasty. Recent evidence demonstrates that laminoplasty preserves posterior elements, reduces postoperative deformity, and minimizes complications. However, there is no standardized framework guiding the choice of approach. We propose a structured preoperative decision model integrating tumor, patient, and surgical factors into a composite biomechanical risk score. Patients with higher scores may benefit from modified laminectomy or fusion, while low-risk patients are ideal candidates for laminoplasty. Additionally, machine-learning models can augment decision-making by predicting individualized deformity risk using imaging and clinical data. This model aims to reduce variability in surgical planning and optimize patient outcomes.