Texture analysis of proximal femur radiographs for osteoporosis assessment

The osteoporosis assessment can be done by observing and analyzing the trabecular pattern in proximal femur. To reduce variability, a machine vision system based on texture analysis using Gabor filter is introduced. Multi-channel filtering implemented by Gabor function, or Gabor filter, is capable to mimics characteristics of the human visual system. In the assessment of osteoporosis, Gabor filter is used to calculate features from trabecular pattern recorded on radiographs of proximal femur. The extracted features represent the quality or structure of the bone, better quality represents better bone strength, lower quality leads to low bone strength and could be suspected as osteoporosis. Extracted features from trabecular pattern recorded in proximal femur radiographs by Gabor filter match with their predetermined Singh index.