Spine lesions are one of the leading causes of disability worldwide, often caused by degenerative changes in the intervertebral disc spaces that harm spinal nerves. Deep Learning tools such as Convolutional Neural Networks have demonstrated effective diagnosis capabilities for spine fractures and lesions; however, intricate medical imaging challenges require further research and innovative approaches. In May 16, 2024, the Radiological Society of North America launched a public competition on Kaggle for the development of Artificial Intelligence methods that assist in the classification and detection of degenerative spine conditions. This challenge offers an MR image dataset consisting of five lumbar spine degenerative conditions, categorized into three degrees of degeneration severity. In this work, we propose a multi-view hybrid quantum deep learning model and its classical counterpart based on a pretrained ResNet18 architecture. The data is split into sagittal T1, axial T2, and sagittal T2/STIR view samples, which are used to train the unimodal models of both approaches for the classes of normal/mild, moderate, and severe. A quantum circuit that functions as a convolutional and pooling layer is introduced as a final feature extraction layer, forming the hybrid quantum model. This hybrid quantum model approach demonstrates superior accuracy, precision, recall, and F1 scores of 0.8433, 0.8749, 0.8433, and 0.8550, respectively, outperforming its classical counterpart across all MR image perspectives. The proposed multi-view hybrid quantum learning approach enhances classification performance, reduces overfitting and loss, and shows promising capabilities in fine feature extraction and in addressing challenges in Computer-Aided Diagnosis (CAD) for the classification of spine degenerative conditions.