Pattern Classification Techniques for the Ultrasonographic Periodtonal Probe

Periodontal disease, commonly known as gum disease, affects millions of Americans. The current method of detecting periodontal disease is painful, invasive, and inaccurate. As an alternative to manual probing, the ultrasonographic periodontal probe is being developed to use RF ultrasound waveforms to measure periodontal pocket depth, which is the main measure of periodontal disease. The methods employed use wavelet transforms and pattern recognition techniques to develop artificial intelligence routines that can automatically detect pocket depth. Results with pattern classification showed that as much as 86.6% of the periodontal pocket depths can be predicted within the manual probe’s 1mm tolerance. Applying ultrasound to dentistry in this way1 is useful for long-term flight situations in the space industry. The same techniques have been applied to several other applications, including using dental ultrasonography to detect cavities in the hard tissue of teeth, indicating flaw severity in pipes, and predicting seismic activity in mines.