Paediatric Bone Age Assessment from Hand X-ray Using Deep Learning Approach

Bone age assessments are methods that doctors use in pediatric medicine. They are used to assess the growth of children by analyzing X-ray images. This work focuses on the development of a deep learning model to estimate from X-ray images. Such a model would avoid the fallacies of subjective methods and raise the accuracy of the assessment. In our work, the model is based on convolutional neural networks (CNN) and is composed of two steps: a preprocessing step generating image masks, and a prediction step that uses these masks to generate the assessment. The model is trained and tested using a public Radiological Society of North America(RSNA) bone age dataset. Finally, experimental results demonstrate the effectiveness of the proposed approach compared to similar works in the literature.