Application of Multi-Spectral Imaging Technique in the Diagnosis of Dental Caries - An In-Vitro Study

Background: Multispectral imaging, combining spectroscopy and imaging, reveals tissue complexities. In dentistry, it distinguishes carious from non-carious teeth by analyzing spectral variations. These variations arise from demineralization and structural changes in enamel and dentin during decay. Objective: To compare the spectra emitted by carious and non-carious teeth and evaluate their differences. Methods: In this in-vitro study, spectral images of 20 carious and 20 non-carious extracted teeth were acquired using the Multispectral Snapshot camera and analyzed with custom computer application to evaluate the diagnostic performance. The obtained data was analyzed using SPSS version 22.0. Results: Spectral analysis showed that differential reflectance peaked at 600 nm ( P < 0.05). However, a two-sample t -test revealed no significant difference in mean reflectance between the groups ( P = 0.7355). While the receiver operating characteristic (ROC) curve’s AUC of 0.43 indicates limited accuracy, the method had a sensitivity of 0.75. These results, though modest, suggest a potential for caries detection and warrant further investigation at higher wavelengths. Conclusion: Near-infrared (NIR) light, particularly at 600 nm, holds significant potential to enhance early caries detection when integrated with advanced imaging techniques.