TY - JOUR
AU - 
AB - J Digit Imaging (2017) 30:427–441 DOI 10.1007/s10278-017-9955-8 1 1 1 1 Hyunkwang Lee & Shahein Tajmir & Jenny Lee & Maurice Zissen & 1 1 1 1 Bethel Ayele Yeshiwas & Tarik K. Alkasab & Garry Choy & Synho Do Published online: 8 March 2017 The Author(s) 2017. This article is published with open access at Springerlink.com Abstract Skeletal maturity progresses through discrete phases, 98.11% of the time. Male test radiographs were assigned a fact that is used routinely in pediatrics where bone age assess- 94.18% within 1 year and 99.00% within 2 years. Using the ments (BAAs) are compared to chronological age in the evalu- input occlusion method, attention maps were created which re- ation of endocrine and metabolic disorders. While central to veal what features the trained model uses to perform BAA. many disease evaluations, little has changed to improve the These correspond to what human experts look at when manually tedious process since its introduction in 1950. In this study, we performing BAA. Finally, the fully automated BAA system was propose a fully automated deep learning pipeline to segment a deployed in the clinical environment as a decision supporting region of interest, standardize and preprocess input 
TI - Fully Automated Deep Learning System for Bone Age Assessment
JF - Journal of Digital Imaging
DO - 10.1007/s10278-017-9955-8
DA - 2017-03-08
UR - https://www.deepdyve.com/lp/unpaywall/fully-automated-deep-learning-system-for-bone-age-assessment-0ArXWB0809
DP - DeepDyve
ER -