eduzhai > Applied Sciences > Engineering >

Deep Learning to Quantify Pulmonary Edema in Chest Radiographs

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 39 pages

Abstract: Purpose: To develop a machine learning model to classify the severity gradesof pulmonary edema on chest radiographs.Materials and Methods: In this retrospective study, 369,071 chest radiographsand associated radiology reports from 64,581 (mean age, 51.71; 54.51 women)patients from the MIMIC-CXR chest radiograph dataset were included. Thisdataset was split into patients with and without congestive heart failure(CHF). Pulmonary edema severity labels from the associated radiology reportswere extracted from patients with CHF as four different ordinal levels: 0, noedema; 1, vascular congestion; 2, interstitial edema; and 3, alveolar edema.Deep learning models were developed using two approaches: a semi-supervisedmodel using a variational autoencoder and a pre-trained supervised learningmodel using a dense neural network. Receiver operating characteristic curveanalysis was performed on both models.Results: The area under the receiver operating characteristic curve (AUC) fordifferentiating alveolar edema from no edema was 0.99 for the semi-supervisedmodel and 0.87 for the pre-trained models. Performance of the algorithm wasinversely related to the difficulty in categorizing milder states of pulmonaryedema (shown as AUCs for semi-supervised model and pre-trained model,respectively): 2 versus 0, 0.88 and 0.81; 1 versus 0, 0.79 and 0.66; 3 versus1, 0.93 and 0.82; 2 versus 1, 0.69 and 0.73; and, 3 versus 2, 0.88 and 0.63.Conclusion: Deep learning models were trained on a large chest radiographdataset and could grade the severity of pulmonary edema on chest radiographswith high performance.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×