Data drivEn COVID19 DEtection with machine learning
DECODE - Data drivEn COVID19 DEtection with machine learning
DECODE is a symptom checker tool to assist patients and family doctors in preliminary diagnosis and early detection of SARS-CoV-2. It uses a machine learning model to check symptoms and provides information that could be helpful to determine whether you need to be tested for SARS-CoV-2.
This service is a result of the common project started at Faculty of Automatic Control, Electronics and Computer Science, Silesian University of Technology and Department of Infectious Diseases and Hepatology, Medical University of Silesia. DECODE machine learning model has been trained on more than 1401 surveys of patients admitted to the hospital with suspected SARS-CoV-2 infection. After admission, patients were asked to fill out the survey about their symptoms and examined with a PCR-RT test to confirm SARS-CoV-2 infection. Based on the survey information and test results we have trained the DECODE model and identified the symptoms that are most related to positive SARS-CoV-2 test outcome.
Please note that this service is not intended to be a substitute for professional medical advice, diagnosis or treatment, and does not constitute medical or other professional advice. If this is a medical emergency, please immediately call emergency personnel (112) to get prompt medical attention.
If you have the results of the PCR test, you can help us to improve the service by providing your symptoms and the test result. We do not ask that you provide any personally identifiable information. We do not collect or store any personal information.
Authors: Małgorzata Bach, Aleksandra Gruca, Paweł Foszner, Joanna Henzel, Jerzy Jaroszewicz, Mateusz Kania, Michał Kozielski, Justyna Mika, Anna Papież, Joanna Polańska, Marek Sikora, Joanna Tobiasz, Aleksandra Werner, Joanna Żyła.