FEATURE EXTRACTION OF LONG BONE (TIBIA) USING TWIST KERNEL INVARIANT DISPARITIES(TKID)

Abstract

Orthopaedic treatment requires proper classification of bone fractures. This helps doctors to assess the injury and plan the treatment accordingly. Different bones have different shapes, sizes and features. Therefore, fractures vary depending on the type, location and pattern of the bone.. Before classification accurate feature extraction is most important to improve the accuracy. This research proposed a new Twist Kernel Invariant Disparities(TKID) method to extract features after segmentation. This research extracts the following features are identified from this segmented image: Mean, SD, Entropy, RMS, Variance, Smoothness, Kurtosis, Skewness, IDM, Contrast, Correlation, Energy, Homogeneity, colour, size, shape, dimensions, convex and concave