As the field of tissue engineering develops, researchers are faced with a large number of degrees of freedom regarding the choice of material, architecture, seeding, and culturing. To evaluate the effectiveness of a tissue-engineered strategy, histology is typically done by physically slicing and staining a construct (crude, time-consuming, and unreliable). However, due to recent advances in high-resolution biomedical imaging, microcomputed tomography (μCT) has arisen as a quick and effective way to evaluate samples, while preserving their structure in the original state. However, a major barrier for using μCT to do histology has been its inability to differentiate between materials with similar X-ray attenuation. Various contrasting strategies (hardware and chemical staining agents) have been proposed to address this problem, but at a cost of additional complexity and limited access. Instead, here we suggest a strategy for how virtual 3D histology in silico can be conducted using conventional μCT, and we provide an illustrative example from bone tissue engineering. The key to our methodology is an implementation of scaffold surface architecture that is ordered in relation to cells and tissue, in concert with straightforward image-processing techniques, to minimize the reliance on contrasting for material segmentation. In the case study reported, μCT was used to image and segment porous poly(lactic acid) nonwoven fiber mesh scaffolds that were seeded dynamically with mesenchymal stem cells and cultured to produce soft tissue and mineralized tissue in a flow perfusion bioreactor using an osteogenic medium. The methodology presented herein paves a new way for tissue engineers to identify and distinguish components of cell/tissue/scaffold constructs to easily and effectively evaluate the tissue-engineering strategies that generate them.