Machine learning-based prediction of COVID-19 diagnosis based on symptoms

  1. Yazeed Zoabi
  2. Shira Deri-Rozov
  3. Noam Shomron

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Effective screening of SARS-CoV-2 enables quick and efficient diagnosis of COVID-19 and can mitigate the burden on healthcare systems. Prediction models that combine several features to estimate the risk of infection have been developed. These aim to assist medical staff worldwide in triaging patients, especially in the context of limited healthcare resources. We established a machine-learning approach that trained on records from 51,831 tested individuals (of whom 4769 were confirmed to have COVID-19). The test set contained data from the subsequent week (47,401 tested individuals of whom 3624 were confirmed to have COVID-19). Our model predicted COVID-19 test results with high accuracy using only eight binary features: sex, age ≥60 years, known contact with an infected individual, and the appearance of five initial clinical symptoms. Overall, based on the nationwide data publicly reported by the Israeli Ministry of Health, we developed a model that detects COVID-19 cases by simple features accessed by asking basic questions. Our framework can be used, among other considerations, to prioritize testing for COVID-19 when testing resources are limited.

Article activity feed

  1. Version published to 10.1038/s41746-020-00372-6
  2. ScreenIT

    SciScore for 10.1101/2020.05.07.20093948: (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
    We used the gradient-boosting predictor trained with the LightGBM 15 Python package.
    Python
    suggested: (IPython, RRID:SCR_001658)

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    We relied on the data reported by the Israeli Ministry of Health, which has limitations and biases. For instance, symptom reporting was more comprehensive in the positive test result group and validated with a directed epidemiological effort 21. This can be reflected by the percentage of COVID-19 positive patients from the overall individuals positive for each symptom, with which we identified features with biased reporting (headache 96.2%, sore throat 92.3% and shortness of breath 92.4%) and symptoms with balanced reporting (cough 27.4% and fever 45.9%). We should also note that all symptoms were self-reported, and a negative value for a symptom can also mean that the symptom was not reported. If we train and test our model while filtering out symptoms of high bias in advance, we get an auROC of 0.862 with a slight change in the SHAP summary plot (Supplementary Figure 1). However, we hope that readers will appreciate the rapid rate at which the pandemic scenario has evolved over the past weeks and understand the limitations of this research while also acknowledging that unusual times call for unusual solutions. We highlight the need for more robust data to complement our framework while also acknowledging the fact that self-reporting of symptoms is always subject to bias. As the COVID-19 pandemic progresses, it is crucial for public organizations and associations to continue recording and sharing robust data with the scientific community that is eager to contribute to the on...

    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.
    • No funding statement was detected.
    • 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 10.1101/2020.05.07.20093948v2 on medRxiv
  4. Version published to 10.1101/2020.05.07.20093948v1 on medRxiv