Fractal-based image texture analysis of trabecular bone architecture

Area of Science:

• Biomedical Imaging
• Materials Science
• Quantitative Bone Histology

Background:

• Trabecular bone's anisotropic structure is crucial for skeletal integrity.
• X-ray imaging is a common modality for bone assessment.
• Accurate characterization of bone texture and anisotropy is vital for understanding bone health and disease.

Purpose of the Study:

• To investigate fractal-based image analysis for extracting textural features of trabecular bone anisotropy from X-ray images.
• To compare the efficacy of power spectrum, Minkowski dimension, and mean intercept length methods.
• To determine the most suitable method for characterizing trabecular bone anisotropy in thick specimens.

Main Methods:

• Fractal-based image analysis techniques were applied to X-ray images of bone specimens.
• Quantification of image textural features using power spectrum, Minkowski dimension, and mean intercept length.
• Global fractal dimension was used to describe image texture roughness.
• Fabric ellipse characterized anisotropic features, with orientation and eccentricity reflecting textural anisotropy.

Main Results:

• Minkowski dimension provided more accurate and consistent estimation of global fractal dimension on synthetic images.
• Minkowski dimension demonstrated higher sensitivity to textural orientation changes in bone X-ray images.
• The eccentricity range tested for bone X-ray images was 0.25-0.80.

Conclusions:

• Minkowski dimension is a more effective measure for characterizing trabecular bone anisotropy in X-ray images of thick specimens.
• Fractal-based analysis, particularly Minkowski dimension, offers valuable insights into bone structural properties.
• This method enhances the quantitative assessment of bone texture and anisotropy from radiographic data.