Towards a supervised incremental learning system for automatic recognition of the skeletal age

Abstract

In this work, we proposed and developed a simple system to estimate skeletal maturity based in using Active Appearance Models in order to create an increasing set of shape-aligned training images which are incrementally stored and used by a K−NN regression classifier. For that purpose, we designed an original layout of landmarks to be located in representative regions of the radiographical image of the hand. Our results show that is possible to use pixels directly as classification features as long as the training and testing images have been previously aligned in shape and pose.