Degenerative Cervical Myelopathy (DCM) is the most common form of atraumatic spinal cord injury in the world. The only treatment for DCM presently is surgical treatment, though only 50-70% of patients will see significant improvement to their condition afterwards. As surgery itself poses a risk, identifying those which will see improvement is imperative to improving the long-term prognosis of the patient. Many studies have utilized Machine Learning (ML) analyses to predict post-surgical outcomes through the use of clinically-derived metrics, to mixed success. This literature has numerous limitations, using only clinically-derived and demographic metrics, or requiring that images be interpreted by time-intensive manual analysis. We aimed to address these shortcomings by evaluating the ability for ML models to predict post-surgical DCM outcomes, using only metrics which were readily available or attainable using imaging-derived metrics obtained via automated means. To accomplish this, we first did a pilot analyses, testing the capabilities for simple regression-based models to assess a patient’s DCM severity based on clinical, demographic, and imaging-derived metrics. Said analysis showed that the majority of Magnetic Resonance Imaging (MRI) derived metrics were associate with DCM severity (p ≤ 0.05). However, when we attempted to developed regression models to predict DCM severity trained on these metrics alone, the resulting models were insufficient, reaching a peak Receiver Operating Characteristic Area Under Curve (ROC-AUC) of 0.713. We then moved on to test a broader assortment of ML techniques for predicting post-surgical DCM outcomes. This revealed that regression-based models trained on metrics derived from T2-weighted axial MRI sequences were most effective, achieving a peak balanced accuracy of 67%. These models appear to utilize an error within the algorithms used for deriving imaging metrics which only occurred around regions of Spinal Cord (SC) compression, resulting in outliers which the ML models could detect. Taken together, it is clear that utilizing both imaging and clinical metrics together alongside ML techniques can produce models capable of predicting the prognostic outcomes of DCM patients post-surgery, though significant strides in model performance remain to be made.