Related Experiment Video
Updated: Jun 25, 2025
07:29
Author Spotlight: Advanced Techniques for Characterizing Tissue Mineralization in Bone Regeneration Research
Published on: September 27, 2024
719
Characterization and quantification of in-vitro equine bone resorption in 3D using μCT and deep learning-aided
Debora M Grass1, Gwladys Malek1, Hubert M Taïeb2
1Comparative Orthopaedic Research Laboratory, Department of Clinical Sciences, Faculty of Veterinary Medicine, University of Montreal, 3200 Sicotte, St-Hyacinthe, QC J2S 2M2, Canada.
Bone
|May 22, 2024
Summary
Researchers developed a 3D imaging method to quantify equine osteoclast resorption volume. This technique accurately measures bone resorption, correlating well with biomarkers and 2D measurements, offering a reliable assessment of osteoclast activity.
Area of Science:
- Biomedical Engineering
- Orthopedics
- Cell Biology
Background:
- High cyclic strain in bone causes microcracks, initiating osteoclast-mediated resorption and bone remodeling.
- Racehorse bone is a valuable model for studying high-intensity loading effects, including microcrack formation and resorption.
- Current in vitro resorption volume assessment often relies on indirect 2D measurements, limiting accuracy.
Purpose of the Study:
- To develop an accurate, high-throughput 3D method for quantifying equine osteoclast resorption volume using micro-computed tomography (μCT).
- To train a convolutional neural network (CNN) for identifying resorption events directly from post-culture μCT images.
Main Methods:
- Equine osteoclasts were cultured on equine bone slices, with μCT imaging performed pre- and post-culture.
- Individual resorption events were isolated and analyzed in 3D to calculate modal volume, maximum depth, and aspect ratio.
- A CNN (U-Net-like) was trained to detect resorption events using post-culture μCT images and validated against known resorption areas and CTX-I biomarker levels.
Main Results:
- 3D resorption volume measurements showed strong correlations with CTX-I levels (p < 0.001) and 2D area measurements (p < 0.001).
- Analysis revealed a spectrum of resorption event shapes, challenging previous 'pit' and 'trench' classifications.
- Simpler, rounded resorption morphologies were most common, with complexity increasing with resorption extent.
Conclusions:
- The developed 3D μCT method provides an accurate and high-throughput quantification of equine osteoclast resorption volume.
- 2D measurements of in vitro osteoclastic resorption remain a robust and reliable proxy for osteoclast activity.
- The study offers new insights into the morphology of bone resorption events.
