Neural network trained to interpret chest x-rays

This week, PLOS Medicine launches our Special Issue on Machine Learning in Health and Biomedicine.
share Share

Researchers in the United States trained a convolutional neural network (CNN) – a deep learning technique – to detect signs of pneumonia in chest x-rays from three large U.S. hospital systems, individually and in combination. Overall, the researchers found that the CNN models showed a significant reduction in accuracy when tested on data from outside the training set. 

Eric Karl Oermann of the Department of Neurological Surgery, Icahn School of Medicine, New York, and co-researchers attempted to validate the trained models' performance internally (with held-out test data) and externally (using test data from a different hospital system). 

The research team reported these key findings:

The CNN trained on data from Mount Sinai Hospital (MSH; 42,396 radiographs from 12,904 patients) had an area under the receiver operating characteristic curve (AUC) of 0.802 (95% confidence interval 0.793-0.812) in held-out internal validation; an AUC of 0.717 (95% CI 0.687-0.746) in data from the National Institutes of Health Clinical Center (NIH; 112,120 radiographs from 30,805 patients), and an AUC of 0.756 (95% CI 0.674-0.838) in data from Indiana University Network for Patient Care (IU; 3,807 radiographs from 3,683 patients). 

Due to limitations of the datasets, and because the features and interrelationships by which CNNs predict outcomes are not easily reduced to simpler, familiar terms, Oermann and co-authors cannot fully assess what factors other than disease prevalence might have led to reduced performance in external validation. The authors say hospital system and department characteristics may have contributed to confounding.

Nonetheless, the study provides evidence that estimates of real-world CNN performance based on held-out internal test data can be overly optimistic.

Source: PLOS

Image Credit: StockSnap, Pixabay

«« Affidea celebrates International Day of Radiology – by looking to the future

Is walk-in CT better than an appointment system? »»


Zech JR, Badgeley MA, Liu M, Costa AB, Titano JJ, Oermann EK (2018) Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study. PLoS Med 15(11): e1002683.

Published on : Wed, 7 Nov 2018

Related Articles

In a study presented at the annual meeting of the Radiological Society of North America (RSNA) in Chicago, researchers using... Read more

The tight roped tug of war concerning artificial intelligence and exponential technologies like deep learning and machine learning... Read more

Radiology and medical imaging is continuously experiencing rapid enhancements including quality of patient care as a result of... Read more

United States, deep learning, convolutional neural network, CNN Neural network trained to interpret chest x-rays

No comment

Please login to leave a comment...

Highlighted Products