Lower back pain might arise due to several issues affecting any structures inside the lumbar spine. The compilation of a medical diagnosis is essential for healthcare professionals to provide appropriate therapy for those with low back pain. Machine learning models utilised in the medical domain for illness diagnosis provide valuable assistance to medical professionals in identifying diseases at an early stage, by analysing symptoms. The objective of this study is to examine and determine the most significant physical characteristics that contribute to spinal malformations, as well as to predict spinal malformation based on the physical data collected from the spine. Various machine learning approaches, like Naive Bayes, SMO, J48, and Bagging, are investigated for the purpose of identifying spinal anomalies. The models are evaluated using a dataset consisting of 310 samples obtained from the Kaggle repository. Various metrics, including as accuracy, mean absolute error (MAE), root mean square error (RMSE) precision, recall, F1 score, are employed to assess the efficacy of classifying spinal patients as either faulty or healthy. The highest accuracy of 84.52% is achieved by Bagging in 20-fold cross validation.