Studies of the Quantitative Analysis of Temporomandibular Joint Sounds Using Spectral Analysis: Part 2. Quantizing for Spectral Analysis

The method of quantitative frequency analysis for TMJ sounds was studied. Since the frequency analysis was strongly affected by some analyzing conditions, these conditions were studied. The subjects were divided into groups according to the pathological state and TMJ sound, and the analyzing conditions for differentiating the groups were selected. The results were as follows: 1. It was necessary to apply over 6 kHz of sampling frequency to enhance the potential for differentiating TMJ sounds. 2. There were no specific relationships between spectrum patterns, pathological states and TMJ sounds. 3. A spectrum peak for TMJ sounds was present in the bass register, but frequency components tended to decrease with higher pitched sounds. Therefore, it is suggested that TMJ sounds should be evaluated with 2 parameters-(a) maximum frequency, and (b) FFT integrated value. 4. As for optimal analyzing conditions, amplitude spectrum, unused window, and 3 mV cut off level should be combined.