Detection of Temporomandibular Joint Disfunction in Juvenile Idiopathic Arthritis Through Infrared Thermal Imaging and a Machine Learning Procedure

Juvenile idiopathic arthritis (JIA) represents the most common rheumatologic disease in childhood, often characterized by temporomandibular joint (TMJ) disfunction (TMD). The gold standard to diagnose this pathology relies on magnetic resonance imaging. Alternatively, electromyographic (EMG) recordings could provide an early and immediate detection of TMD. Particularly, temporalis muscle is known to exhibit a greater EMG activity more frequently than the masseter in pathological subjects. Since muscular activity may influence the superficial circulation and, consequently, the skin temperature, the capabilities of functional thermal infrared imaging (fIRI) to detect TMD were also investigated. In this study, the feasibility of a multivariate data-driven approach based on General Linear Model to estimate the EMG ratio between masseter and temporalis (sEMG-M/T) from fIRI features was investigated. A leave-one-subject-out cross-validation was implemented to test the generalization capability of the model (r = 0.55; p = 1.72.10(-6)). Moreover, the output of the model was used to classify TMD and healthy controls. Since the two classes were unbalanced, a bootstrap procedure was applied. The performances of the classifier were investigated through Receiver Operating Characteristic analysis, which exhibited an area under the curve of 0.71. The results suggested that fIRI could be a relative cheap and simple to use tool for TMD assessment.