Skip to main content
. Author manuscript; available in PMC: 2021 Sep 1.
Published in final edited form as: J Thorac Imaging. 2020 Sep;35(5):285–293. doi: 10.1097/RTI.0000000000000505

Figure 4: Receiver-operator characteristic (ROC) curve for the performance of U-Net segmentation for detection of pneumonia.

Figure 4:

Including all radiographs (blue), overall performance yielded an area under the curve (AUC) of 0.854, which corresponds an accuracy of 81.6%, sensitivity of 82.8%, specificity of 72.6%, positive predictive value of 47.9%, and negative predictive value of 93.3%. Excluding radiographs with other diagnoses (not pneumonia), AUC was 0.944. Excluding normal radiographs, AUC was 0.788.