A high-throughput semi-automated bone segmentation workflow for murine hindpaw micro-CT datasets

Area of Science:

• Biomedical Imaging
• Skeletal Biology
• Quantitative Analysis

Background:

• Micro-computed tomography (μCT) is crucial for longitudinal bone volume assessment in various conditions.
• Manual segmentation of complex murine hindpaws is time-consuming, limiting comprehensive analysis.
• A high-throughput, user-friendly, semi-automated method is needed for murine hindpaw μCT datasets.

Purpose of the Study:

• To develop and validate a semi-automated segmentation method for murine hindpaw μCT data.
• To improve the efficiency and reliability of bone volume quantification.
• To facilitate pre-clinical research in bone and joint analysis.

Main Methods:

• Longitudinal in vivo μCT scans of C57BL/6 mice (male and female).
• Ex vivo μCT of hindpaws to assess resolution and integration time effects.
• Watershed-based semi-automated segmentation in Amira software with user correction.

Main Results:

• Semi-automated segmentation yielded comparable bone volumes to manual methods.
• Significant reduction in segmentation time for both experienced and novice users.
• Excellent inter-user reliability (ICCs >0.9) for most bone structures.

Conclusions:

• The semi-automated approach offers substantial reliability and throughput advantages.
• Standardization of bone volume measures across users and institutions is achievable.
• Accelerates pre-clinical bone and joint research towards clinical translation.