Although the topic of tooth fractures has been extensively analyzed in the dental literature, there is still insufficient information on the potential effect of enamel microcracks (EMCs) to the underlying tooth structures. For precise examination of tooth structure damage in the area of EMCs (i.e. whether it crosses the dentin-enamel junction (DEJ) and reaches dentin or pulp), volumetric (three-dimensional (3D)) evaluation of EMCs is necessary. The aim of this study was to present an X-ray micro-computed tomography (μCT) as a technique suitable for 3D non-destructive visualization and qualitative analysis of different severity teeth EMCs. Extracted human maxillary premolars were examined using a μCT instrument ZEISS Xradia 520 Versa. In order to separate (segment) cracks from the rest of the tooth a Deep Learning Tool was utilized within the ORS Dragonfly software. The scanning technique used allowed for the recognition and detection of EMCs that are not only visible on the outer surface but also those that are deeply buried inside the tooth. The 3D visualization combined with Deep Learning segmentation enabled evaluation of EMC dynamics as it extends from the cervical to the occlusal part of the tooth, and precise examination of EMC position with respect to the DEJ.