/covidpred

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

A COVID-19 Prediction Model Using Symptoms

From: Machine learning-based prediction of COVID-19 diagnosis based on symptoms, npj Digital Medicine; doi:10.1038/s41746-020-00372-6

Previously: COVID-19 diagnosis prediction by symptoms of tested individuals: a machine learning approach, medRxiv; doi:10.1101/2020.05.07.20093948

Model Predictors and Exact Variable Names (True = 1, False = 0)

  • Age over 60 - Age_60
  • Sex - Male (Male=1, Female=0)
  • Cough - Cough
  • Shortness of breath - Shortness_of_breath
  • Fever - Fever
  • Sore throat - Sore_throat
  • Headache - Headache
  • Contact with a confirmed individual - Contact_with_confirmed

Model Outcome

The probability of being diagnosed with a COVID-19 infection.

Use

  1. Import lgbm_model_*.txt using LightGBM 2.3.1 on Python 3.6.

  2. Predict using your data.

Files in this repository

  • lgbm_model_all_features.txt - The predictor that uses all 8 features
  • lgbm_model_balanced_features.txt - The predictor that uses only balanced symptoms
  • hyperparameters.txt - The hyperparameters used by lightGBM
  • data/corona_tested_individuals_ver_0083.english.csv.zip - The tested individuals dataset downloaded from https://data.gov.il/dataset/covid-19 on November 15, 2020 and translated into English
  • data/corona_tested_individuals_ver_006.english.csv.zip - The tested individuals dataset downloaded from https://data.gov.il/dataset/covid-19 on May 4, 2020 and translated into English. This is the version we used for the analysis.