Blind separation of temporomandibular joint sound signals

In order to develop a cheap, efficient and reliable diagnostic tool for the detection of temporomandibular joint disorders (TMD), sounds from the temporomandibular joint (TMJ) are recorded using a pair of microphone inserted in the auditory canals. However, the TMJ sounds originating from one side of the head can also be picked up by a microphone at the other side. Blind source separation (BSS) is thus proposed as a method to recover the original sound. The authors propose to use non-casual filters for the separation of TMJ signals. The algorithm is based on information theory and is an extension of early work by Torkkola (see IEEE workshop on Neural Networks for Signal Processing, Kyoto, Japan, 1996). The separation was successful and the output can now be used for subsequent analysis of TMJ sounds