Estimation of Compressive Strength of Femur Bones using Radiographic Imaging and Spectral Analysis

Abstract

In this work, the assessment of the mechanical strength of human femur trabecular bone and its classification into normal and abnormal images are carried out using digital image processing and Fourier transform methods. Normal and abnormal femur trabecular images (N=20) were recorded using planar radiography with standard image acquisition protocol. From the images, the primary strength component of the trabeculae are delineated using image processing procedures and are then subjected to Fourier transform analysis. The zero (DC), First (FMO), and Second moments of inertia (SMO) are the parameters considered and are correlated with presence and absence of mineralization in the image. The results demonstrate correlation between mineralization and FMO in normal and abnormal cases. Distinct variation in SMO between normal and abnormal was observed. It appears that SMO is the useful parameter to assess the mechanical strength of femur trabeculae. In this paper the objectives, methodology, significance results and the conclusions are presented.