Researchers at Stanford University have demonstrated that a deep learning convolutional network model can estimate bone age as well as human radiologists.
The team says the results suggest significant potential for deep learning models for diagnostic imaging without requiring specialized subject matter knowledge or image-specific software engineering.
The researchers divided 14,036 left-hand radiographs into two groups, with the first set using a mean of bone age estimates from the clinical report and three additional human reviewers as the reference standard; the second set contained 913 unidentified images. The researchers assessed the deep learning model's performance by comparing the root mean square and mean absolute difference between the model estimates and the reference standard bone ages.
The estimates of the model, clinical report, and three reviewers were 95% in agreement.
From Health Imaging
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA