Prolonged standing is a common everyday activity in many jobs and daily routines, frequently resulting in musculoskeletal disorders, primarily the spine. The human spine is a complex structure which goes into supporting the body’s weight and for doing all possible movements. Yet prolonged periods of standing can put considerable stress on the spine and thereby create the uncomfortable and painful, as well as, long-term health problems. Spine stress during prolonged standing plays a major role in occupational health and recognition of the factors contributing to it is fundamental in the development of preventive measures. The goal of this research is to examine the effects, with machine learning models, on the human spine using real-time data and enhance them with stress-related physiological parameters to find the most critical factors in spine health. Real-time data were collected on individuals’ spine health, posture, and other physiological conditions while standing for long periods of time (i.e., columns such as pelvic incidence, pelvic tilt, lumbar lordosis angle, sacral slope, pelvic radius, degree of spondylolisthesis, and class). The data collected were cleaned and preprocessed, and additional stress-related physiological parameters were created using synthetic data (such as heart rate, blood pressure (systolic/diastolic, cortisol levels, respiration rate, skin conductance, and body temperature). Logistic regression, support vector machine (SVM), decision tree, and random forest were developed and trained, and evaluated in terms of accuracy, precision, recall, and F1 score. Results show that an enriched dataset with more stress-related physiology parameters offers a better picture of what plays into spine health. The logistic regression model was found to be the best-performing model, obtaining the highest accuracy, precision, recall, and F1 score with the enriched dataset, with accuracy of 90.32%, 90.32% precision, 90.32% recall, and 90.32% F1 score. By adding stress-related physiological parameters, the model performance was significantly improved and it underscores the importance of including physiological effects in studying spine health under prolonged standing. The performance metrics for the models trained on the original dataset were as follows: The model is trained in logistic regression with report of 85.48% accuracy, 85.28% precision, 85.48% recall, and 85.36% F1 score; support vector machine (SVM) with 80.65% accuracy, 80.65% precision, 80.65% recall, and 80.65% F1 score; decision tree with 77.42% accuracy, 77.42% precision, 77.42% recall and 77.42% F1 score; and random forest with 77.42% accuracy, 76.32% precision, 77.42% recall, and 76.50% F1 score. Results were significantly better on the enriched dataset, showing that parameterizing with physiological data provided information beyond the raw hand motion. Artificial intelligence machine learning models analysis of ongoing effects on the human spine can also identify the major determinants of spine health, as research shows. Thus, future work could collect more data, try other machine learning algorithms, and build customized recommendation systems for users. In addition, the results of this study can be used to direct the development of preventive measures and interventions directed at reducing the negative consequences of long standing on spine health and to assist add to the existing knowledge corpus in occupational health and safety.