The COVIDiag is based on ensemble learning, was trained by 488 patients (244 COVID-19 and 244 non-CoOVID19), was tested by 124 patients (62 COVID-19 and 62 non-CoOVID-19), and was validated by 450 patients from five Countries (Argentina, Turkey, Iran, Italy, and The Netherlands).
Install the COVIDiag software and run it. You need MATLAB Runtime 2017a to run the COVIDiag software. The below figure shows the environment of the COVIDiag software.
Descriptions of the steps of using the software are as follows:
- Write the name, ID of the patient, and date.
- Determine the CT findings of the patient. Note that all the CT findings should be filled.
- Click on the “Check” icon and wait for the result. The result will appear in the “Result” icon.
- You can save the name, ID, date, CT findings, and the result by clicking on the “Record” button. Before that, you should define a directory path in which the output should be saved.
- Click on “Clear” and diagnose the next patient.
Please Cite the following articles:
- Ardakani AA, Acharya UR, Habibollahi, Mohammadi A. COVIDiag: a clinical CAD system to diagnose COVID-19 pneumonia based on CT findings. European Radiology. 2021;31:121–130. (DOI: https://doi.org/10.1007/s00330-020-07087-y)
- Ardakani AA, Kwee, R. M., Mirza-Aghazadeh-Attari, M., Castro, H. M., Kuzan, T. Y., Altintoprak, K. M., … & Mohammadi, A. A practical artificial intelligence system to diagnose COVID-19 using computed tomography: A multinational external validation study. Pattern Recognition Letters. 2021;152:42-49. (DOI: https://doi.org/10.1016/j.patrec.2021.09.012)