Deep learning model to predict the need for mechanical ventilation using chest X-ray images in hospitalised patients with COVID-19

  1. Anoop R Kulkarni
  2. Ambarish M Athavale
  3. Ashima Sahni
  4. Shashvat Sukhal
  5. Abhimanyu Saini
  6. Mathew Itteera
  7. Sara Zhukovsky
  8. Jane Vernik
  9. Mohan Abraham
  10. Amit Joshi
  11. Amatur Amarah
  12. Juan Ruiz
  13. Peter D Hart
  14. Hemant Kulkarni

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

There exists a wide gap in the availability of mechanical ventilator devices and their acute need in the context of the COVID-19 pandemic. An initial triaging method that accurately identifies the need for mechanical ventilation in hospitalised patients with COVID-19 is needed. We aimed to investigate if a potentially deteriorating clinical course in hospitalised patients with COVID-19 can be detected using all X-ray images taken during hospitalisation.

Methods

We exploited the well-established DenseNet121 deep learning architecture for this purpose on 663 X-ray images acquired from 528 hospitalised patients with COVID-19. Two Pulmonary and Critical Care experts blindly and independently evaluated the same X-ray images for the purpose of validation.

Results

We found that our deep learning model predicted the need for mechanical ventilation with a high accuracy, sensitivity and specificity (90.06%, 86.34% and 84.38%, respectively). This prediction was done approximately 3 days ahead of the actual intubation event. Our model also outperformed two Pulmonary and Critical Care experts who evaluated the same X-ray images and provided an incremental accuracy of 7.24%–13.25%.

Conclusions

Our deep learning model accurately predicted the need for mechanical ventilation early during hospitalisation of patients with COVID-19. Until effective preventive or treatment measures become widely available for patients with COVID-19, prognostic stratification as provided by our model is likely to be highly valuable.

Article activity feed

  1. Version published to 10.1136/bmjinnov-2020-000593
  2. ScreenIT

    SciScore for 10.1101/2020.08.17.20176917: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Network Training and Validation: We used the Tensorflow 2.2.0 (https://www.tensorflow.org/) and Keras 2.3.0-tf framework (https://keras.io/) for model training and evaluation.
    https://www.tensorflow.org/
    suggested: (tensorflow, RRID:SCR_016345)
    The Jupyter notebook congaing all the Python code is available with the authors and will be shared upon receipt of reasonable request.
    Python
    suggested: (IPython, RRID:SCR_001658)

    Results from OddPub: Thank you for sharing your code.

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    The results of our study should be considered in the light of some limitations. First, this was a retrospective, observational evaluation and the confounding and bias implicit in such an investigation will remain a limitation. Second, the data for this study were derived from a single center and the generalizability of this approach to other settings needs to be established in further studies. Third, we restricted our model to the use of chest radiographs only. However, additional clinical parameters at the time of hospital admission such as respiratory rate, oxygenation status (e.g. the ROX index[20]) and altered mental status [21] along with socio-demographic characteristics, comorbidity profile and laboratory investigations can potentially further improve the prediction. Future studies need to evaluate these possibilities, but our focus was to use an objective measure such as a chest radiograph and provide a tool to the critical care provider with a reasonable expectation of the future course of disease in a given patient.

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  3. Version published to 10.1101/2020.08.17.20176917v1 on medRxiv