Originally published online:
https://doi.org/10.1148/radiol.2018180921
Erratum in:
Radiology 2019;291(1):272 DOI:10.1148/radiol.2019194005
There were some errors in an early online version.
In the abstract Results: “Normal chest radiographs were detected by our AI system with a sensitivity of 71%, specificity of 95%, PPV of 73%, and NPV of 99%” should read “ Normal chest radiographs were detected by our AI system with a sensitivity of 71%, specificity of 95%, PPV of 73%, and NPV of 94%.”
In Results, third line under “Deep Learning Architecture for Criticality Prediction from Image Data,” the sentence “AI performance was good, with a sensitivity of 71%, specificity of 95%, PPV of 73%, and NPV of 99% for normal radiographs (Fig 4) and a sensitivity of 65%, specificity of 94%, PPV of 61%, and NPV Automated Triaging of Adult Chest Radiographs with Deep Artificial Neural Networks
Mauro Annarumma, Samuel J. Withey, Robert J. Bakewell, Emanuele Pesce, Vicky Goh, Giovanni Montana of 99% for critical radiographs” should read “AI performance was good, with a sensitivity of 71%, specificity of 95%, PPV of 73%, and NPV of 94% for normal radiographs (Fig 4) and a sensitivity of 65%, specificity of 94%, PPV of 61%, and NPV of 95% for critical radiographs.”
In Discussion, third line, the sentence “Similarly, our deep CNN–based computer vision system was able to separate normal from abnormal chest radiographs with a sensitivity of 71%, specificity of 95%, and NPV of 99%” should read “Similarly, our deep CNN–based computer vision system was able to separate normal from abnormal chest radiographs with a sensitivity of 71%, specificity of 95%, and NPV of 94%.”
In table 3, the data for NPV should read as follows: 94, 90, 72, and 95.