Deep learning models match expert radiologists in estimating children's bone age
In short: Researchers developed a deep-learning neural network to assess skeletal maturity by analyzing hand X-rays of children. The study found that the model could estimate bone age with an accuracy comparable to that of human experts, showing a mean difference of zero years compared to professional reviewers.
This research explores how artificial intelligence can streamline clinical assessments of growth and development. By training on a large dataset of radiographs from two children's hospitals, the model was tested against both human radiologists and previously existing automated tools.
The study used two test sets to evaluate performance, including one based on published reports from a digital atlas. A limitation of this research is that it compares the model's performance to other models and experts without specifying the specific clinical outcomes influenced by these age estimates.
Why it matters: AI models are becoming increasingly capable of performing complex medical diagnostic tasks with the same precision as human specialists.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Summary by Intellect · source: Research: Larson et al. (2017). Performance of a Deep-Learning Neural Network Model in Assessing Skeletal Maturity on Pediatric Hand Radiographs. Radiology.
To cite this card, link to https://intellectplus.co.uk/app/history-research-deep-learning-models-match-expert-radiologists-in-estimating.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in History · Latest · A–Z · All topics
