The modern approach to the treatment of caries requires maximum preservation of tooth tissues and pulp viability, for which it is necessary to know the residual thickness of dentin during its removal. Currently existing methods (Cone-beam Computed Tomography, electrical impedance device, and optical coherence tomography) are not widely used in clinical practice due to the laboriousness of their use or low accuracy. We evaluated the capabilities of the diffuse reflectance spectroscopy (DRS) method for determining dentine thickness in situ. Dentin tissues transmit light well in the visible and near-IR range, which makes it possible to detect the optical response of the dental pulp. The pulp contains hemoglobin and water, while dentin contains no hemoglobin, and its water content is less than 10%. Thus, the selection of the contributions of these components allows estimating the thickness of the dentin. Our results show a strong correlation (> 0.9) between dentin thickness and the amplitudes of the water and hemoglobin components. However, hemoglobin content is more susceptible to changes, caused by inflammation or the action of anesthesia. Thus, the most promising approach is use the water component as a proxy. The proposed method can be the basis for the development of a fiber-optic laser probe for clinical dentistry.