Automated 3D Mapping and Quantification of Haversian Architecture in Bone Tissue




Betancourt, Alejandro Morales
Montelongo, Sergio A.
Appleford, Mark R.
Guda, Teja

Journal Title

Journal ISSN

Volume Title


Office of the Vice President for Research


Bone tissue has a unique architecture necessitating laborious characterization at various scales. Since the extracellular matrix (ECM) exists in a solid phase, traditional histology reduces the mineral content of ECM to obtain high-resolution 3d micrographs and is operator dependent in terms of analysis. Other imaging techniques such as microCT (μCT), are limited by the spatial resolution of the μCT scanner and provide minimal cellular correlation. The vascularization inside the bone (Haversian system) appears to exist in continually remodeling patterns with bone turnover, making it difficult to numerically assess the Haversian canal architecture. Our novel technique proposes the 3D characterization of bone tissue, by using light microscopy and MATLAB®, to acquire microscopic 3D images of bone vasculature with enhanced spatial resolution. Briefly, the right femurs of eight rats were sectioned in cross sectional slices of 1 mm thickness. The slices were exposed to a Villanueva osteochrome stain protocol followed by sequential dehydration, poly methyl-methacrylate infiltration, and polymerization. Using a combination of a novel, automated focus-adjusting system in conjunction with light microscopy, we obtained 100 sequential discrete images at 1μm steps along the z-stack at 100x in-plane magnification. The images were processed using MATLAB® software, obtaining a 3D representation of the Haversian system. From these 3D images we quantified Haversian number, and vessel parameters, including surface, volume, channel thickness and porosity ratio to tissue volume. Data analysis shows differences in the Haversian thickness in the three regions studied, showing that the overall vessel volume remains relatively conserved, but the Haversian architecture changes further away from the proximal region. Due to optical microscope limitations, this technique only can measure samples up to 100 microns in depth. This protocol can potentially be used to obtain 3D architectural quantification of other tissues by modifying staining protocols.



Haversian, three-dimensional, cortical bone