3
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: not found
      • Article: not found

      Assessment of Convolutional Neural Networks for Automated Classification of Chest Radiographs

      Read this article at

      ScienceOpenPublisherPMC
      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Purpose To assess the ability of convolutional neural networks (CNNs) to enable high-performance automated binary classification of chest radiographs. Materials and Methods In a retrospective study, 216 431 frontal chest radiographs obtained between 1998 and 2012 were procured, along with associated text reports and a prospective label from the attending radiologist. This data set was used to train CNNs to classify chest radiographs as normal or abnormal before evaluation on a held-out set of 533 images hand-labeled by expert radiologists. The effects of development set size, training set size, initialization strategy, and network architecture on end performance were assessed by using standard binary classification metrics; detailed error analysis, including visualization of CNN activations, was also performed. Results Average area under the receiver operating characteristic curve (AUC) was 0.96 for a CNN trained with 200 000 images. This AUC value was greater than that observed when the same model was trained with 2000 images (AUC = 0.84, P < .005) but was not significantly different from that observed when the model was trained with 20 000 images (AUC = 0.95, P > .05). Averaging the CNN output score with the binary prospective label yielded the best-performing classifier, with an AUC of 0.98 (P < .005). Analysis of specific radiographs revealed that the model was heavily influenced by clinically relevant spatial regions but did not reliably generalize beyond thoracic disease. Conclusion CNNs trained with a modestly sized collection of prospectively labeled chest radiographs achieved high diagnostic performance in the classification of chest radiographs as normal or abnormal; this function may be useful for automated prioritization of abnormal chest radiographs. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by van Ginneken in this issue.

          Related collections

          Author and article information

          Journal
          Radiology
          Radiology
          Radiological Society of North America (RSNA)
          0033-8419
          1527-1315
          February 2019
          February 2019
          : 290
          : 2
          : 537-544
          Affiliations
          [1 ]From the Departments of Computer Science (J.A.D., C.R.), Biomedical Data Science (D.Y., D.L.R.), and Radiology (C.P.L., D.L.R., M.P.L.), Stanford University, 300 Pasteur Dr, Stanford, CA 94305.
          Article
          10.1148/radiol.2018181422
          6358056
          30422093
          6711a0f9-bae5-474a-a4a3-69534214f3a5
          © 2019
          History

          Comments

          Comment on this article