Comparing machine learning algorithms to predict COVID‑19 mortality using a dataset including chest computed tomography severity score data

A total of 6854 suspected cases had been referred to Ayatollah Talleghani Hospital, where records of 815 positive RT-PCR patients remained after applying the exclusion criteria. Overall, 54.85% of the enrolled patients were male and the mean age of the study population was 57.22 ± 16.76 years. As was mentioned, the deceased group contained only 108 records (13%) and SMOTE method has been used to balance these data. The number of records in this class was raised to 707 after balancing the dataset.

Feature selection

Twenty-seven features were chosen as the most important and relevant…