Content Based Image Retrieval: Bone Age Assessment

Bone age assessment using a hand radiograph is an important clinical tool in the area of paediatrics, especially in relation to endocrinological problems and growth disorders. This paper proposes Standard Deviation based skeletal Bone Age Assessment based on features extracted from phalanx. The system works according to the renowned Tanner and Whitehouse method based on the Phalanx Region of Interest. The system ensures accurate and robust bone age assessment of children between the age group 0-10 years. The proposed work focuses on pre-processing stage in which cropping the phalanx ROI from the input image by followed by filter is done. This follows morphological feature extraction during which three geometric features are extracted, one from each of the phalanx leading to a set of 14 features. The extracted features are used to train the system in the beginning till end to estimate the bone age. The system is tested on a set of radiographs and the results are discussed.