Incorporated region detection and classification using deep convolutional networks for bone age assessment

作者：

Highlights：

• The first attempt to integrate TW3 and CNN-based method based on deep learning.

• Explore the expert knowledge from TW3 for bone age assessment by deep convolution networks.

• Incorporation of TW3 scheme with deep learning shows better performance than the GPbased method.

• Achieves a mean absolute error of about 0.59 years between manual radiology expert and automatic evaluation.

摘要

•The first attempt to integrate TW3 and CNN-based method based on deep learning.•Explore the expert knowledge from TW3 for bone age assessment by deep convolution networks.•Incorporation of TW3 scheme with deep learning shows better performance than the GPbased method.•Achieves a mean absolute error of about 0.59 years between manual radiology expert and automatic evaluation.

论文关键词：Bone age assessment,Convolutional neural networks,Tanner-Whitehouse,Greulich and Pyle

论文评审过程：Received 12 September 2018, Revised 3 March 2019, Accepted 27 April 2019, Available online 30 April 2019, Version of Record 23 May 2019.

论文官网地址：https://doi.org/10.1016/j.artmed.2019.04.005