Bone age Prediction

Deep CNN Approaches for predicting bone age from hand X-ray radiographs

Deep ConvolutionalNeural Networks (CNNs) have proven to be a powerful tool for image classification, image regression, machine vision, and feature extraction tasks. They have been applied successfully to a variety of medical imaging tasks, including bone age prediction. This paper explains how we built deep convolutional neural networks to predict bone age in months using hand X-ray image data as input. In order to extract functional results, we used Inception V4, which is a deep convolutional neural network (CNN) architecture for image classification and regression tasks. It was introduced in 2014 by Google researchers and published in 2016. Another approach that is used is ResNet-50, a deep convolutional neural network (CNN) architecture for image classification and machine vision tasks. It was developed by Microsoft researchers and published in 2015. Although both of these approaches have different structures and architectures, they both produce reasonable results.