Currently, it is not possible to accurately predict the prognosis of temporomandibular disorders (TMDs). The objective of this study was to test two hypotheses concerning predicting longitudinal changes in the temporomandibular joint (TMJ) of human subjects. Hypothesis 1: To test if Anatomic-Psychologic Score (APS), an instrument which utilizes easily measured variables, predicted longitudinal changes in TMJ tissue integrity. Hypothesis 2: To test if Machine Learning Models accurately rank ordered anatomic and psychosocial variables that were associated with longitudinal changes in TMJ tissue integrity.