# Novel image registration approach for combining 2D Osterix and collagen bundles images with 3D micro-CT

**Authors:** Mireille Ngokingha Tchouto, Julia Mehl, Saeed Khomeijani Farahani, Daniel Baum, Georg N Duda

PMC · DOI: 10.1093/jbmrpl/ziag009 · 2026-01-28

## TL;DR

This paper introduces a new method to align 2D histology images with 3D micro-CT scans to better study bone healing processes.

## Contribution

A novel 2D-3D image registration workflow is introduced to spatially align histological and μCT data for bone healing analysis.

## Key findings

- The registration method successfully aligned 2D histology and μCT slices across different fixation conditions.
- The framework allows co-localization analysis of OSX, collagen, and hydroxyapatite in a single spatially matched image.
- The approach is consistent and applicable for studying spatial relationships during bone healing.

## Abstract

To study bone healing, different imaging techniques are typically employed. Histological and immunohistological methods allow the visualization of Osterix (OSX) expression and collagen structures, whereas μCT imaging enables the assessment of mineralized tissue. However, a clear spatial alignment between the information obtained from these 2 modalities is still lacking. In this study, we present a technological approach for registering 2D histological sections of collagen bundles and OSX signals with 3D volumetric μCT data. We applied this method to datasets from animals with rigid and semi-rigid fracture fixation to resemble fast and effective healing and delayed healing. Using our 2D-3D registration workflow, we were able to identify corresponding 2D μCT and histological slices, and we showed that the algorithm performed consistently across both fixation conditions, resembling the healing consequences. Thus, we could illustrate how such image registration techniques could be used to study the co-localization of OSX, collagen, and hydroxyapatite (HA). This framework enables the visualization and direct comparison of OSX, collagen, and HA within a single, spatially matched image, providing a tool for future studies to quantitatively explore tissue co-localization and spatial relationships during bone healing.

## Linked entities

- **Genes:** SP7 (Sp7 transcription factor) [NCBI Gene 121340]
- **Proteins:** COL3A1 (collagen type III alpha 1 chain)
- **Chemicals:** hydroxyapatite (PubChem CID 14781)

## Full-text entities

- **Genes:** Sp7 (Sp7 transcription factor 7) [NCBI Gene 170574] {aka 6430578P22Rik, C22, Osx}, Cdh5 (cadherin 5) [NCBI Gene 12562] {aka 7B4, Cd144, VE-Cad, VECD, VEcad, Vec}, SP7 (Sp7 transcription factor) [NCBI Gene 121340] {aka OI11, OI12, OSX, osterix}
- **Diseases:** hematoma (MESH:D006406), fracture (MESH:D050723), BMD (MESH:D020388), analgesia (MESH:D000699), infection (MESH:D007239), hypothermia (MESH:D007035), dislocation (MESH:D004204)
- **Chemicals:** oxygen (MESH:D010100), clindamycin (MESH:D002981), isoflurane (MESH:D007530), HA (MESH:D017886), medetomidine (MESH:D020926), SHG (-), tramadol (MESH:D014147), Temgesic (MESH:D002047), PBS (MESH:D007854), paraformaldehyde (MESH:C003043)
- **Species:** Mus musculus (house mouse, species) [taxon 10090], Homo sapiens (human, species) [taxon 9606]

## Figures

4 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12906291/full.md

---
Source: https://tomesphere.com/paper/PMC12906291