An improved bone age assessment using advanced image processing and deep learning approach / Liang Kim Meng

Pediatricians often apply bone age assessment to measure the skeletal maturity of children and to predict the future height. These discrepancies are good indicators for diagnosing growth disorders. Normally, left hand skeletal is employed in this assessment. The low quality of ossification sites of carpals deteriorates the pediatrician’s visibility in inspecting the pertinent radiographic manifestations. This in turn affects the bone age assessment. Therefore, we have to enhance the quality before assessing them. Histogram equalization is one of the contrast enhancement techniques that suit this type of enhancement. Existing histogram equalizations, however, are confronting with problems in preserving the brightness and details as well as preventing the contrast from being over-enhanced or under-enhanced simultaneously. The comprehensive histogram equalization was proposed by considering all criteria of the desired histogram-equalized image to produce moderately contrast enhanced carpals’ ossification sites. Qualitative results show that the determining features of maturity stages have been emphasized in some of the Pareto optimized image. The improvement for Pareto optimized image by bi-histogram equalization is significant for five stages from stage D to stage H with improvement accuracy of 7.16%, 12.47%, 16.03%, 21.21% and 18.51%, respectively. Findings concluded that the Pareto optimized images able to improve the classifier accuracy that estimate the maturity stage of the carpal bones