Hyperspectral imaging system associated with novel subtracting image processing algorithm for dental caries early detection

Abstract

Early detection of caries is vital for demineralization reverse offering a less pain, as well as accurate and precise caries removal. In this study, the difference in optical properties of normal tissue and human caries lesion has been used for early diagnosis, using laser induced fluorescence spectroscopy. The human tooth sample was illuminated with visible band sources (488, 514 and 633 nm) with power 5 mW. The reflected and emitted light of tested sample was collected using hyperspectral camera in an attempt to generate multispectral images (cubic image). The variation of reflected and emitted power as function in wavelength was employed to generate characteristic spectrum of each tooth tissue. Human teeth caries tissue lesion releases its excess power by emitting fluorescence light producing chemical footprint signature; this signature is dependent on the elemental composition of tooth elements and caries state. This non-invasive, non-contact and non-ionizing imaging system with associated novel pattern recognition algorithm was employed to diagnose and classify dental (enamel and dentin), caries, and white spot lesion. It was reported that the perceived fluorescence emission is function of the illuminating wavelength. While enamel and dentin caries were distinguished and characterized at 633 nm illuminating wavelength; white spot lesion cannot be detected. The enamel caries, dentin caries, and white spot lesion were contoured and recognized at 488 nm and 514 nm wavelengths. Therefore full recognition was achieved through generated cubic image after sample irradiation at 488 nm and 514 nm. This non-invasive, non-contact, and non-ionizing imaging technique with customized image processing algorithm offered high sensitivity, high resolution, and early caries detection with optimum performance at 514 nm.