Automatic segmentation method for bone and blood vessel in murine hindlimb.

Area of Science:

• Biomedical imaging
• Medical image analysis
• Murine hindlimb anatomy

Background:

• Accurate segmentation of murine hindlimb vasculature is crucial for research.
• Existing methods struggle with bone-vessel separation, false positives from vesselness algorithms, and fragmented vessel extraction.

Purpose of the Study:

• To develop an automated method for precise bone and blood vessel segmentation in murine hindlimbs.
• To overcome challenges in distinguishing bone from vessels, handling sharp boundaries, and reconnecting discontinuous vessel segments.

Main Methods:

• Employing the split Bregman method for initial bone segmentation to improve bone-vessel differentiation.
• Utilizing an edge extension strategy with vesselness and multiscale bi-Gaussian filtering to mitigate false positives on sharp boundaries.
• Implementing a centerline and morphological dilation-based method for reconnecting segmented blood vessels.

Main Results:

• The split Bregman method achieved bone segmentation comparable to manual methods, outperforming thresholding.
• The edge extension strategy effectively reduced false positives in nonvascular regions.
• The proposed vascular connection method successfully reconnected broken vessels while preserving intact ones.

Conclusions:

• The integrated method provides accurate segmentation of both bones and blood vessels in murine hindlimbs.
• Split Bregman enhances bone-vessel distinction by considering intensity and geometry.
• The vesselness method with edge extension refines vessel segmentation, and the connection method ensures vessel continuity.