ATHENS – The first results of interdisciplinary, inter-institutional research aimed at early prediction of the outcome of the COVID-19 disease in patients positive for SARS-CoV-2, with the help of Artificial Intelligence, are significant.
The scientific paper entitled “Early prediction of COVID-19 outcome using artificial intelligence techniques and only five laboratory indices” was published in the international scientific journal Clinical Immunology. The study, in which researchers from ten international universities participated, was coordinated by EKPA School of Medicine Professors Styliani Kokkori, Evangelos Terpos and Thanos Dimopoulos (EKPA Chancellor) and Professor Panagiotis Asteris from the Computational Engineering Laboratory of the Higher School of Pedagogy and Technological Education.
In particular, using an innovative methodology, a new index (alpha-index) was proposed by which 25 laboratory parameters at first from 248 consecutive adult COVID-19 patients for database creation, training, and development of ANN models developing a new alpha-index to assess association of each parameter with outcome.
The first five laboratory indices ranked by importance were Neutrophil-to-lymphocyte ratio, Lactate Dehydrogenase, Fibrinogen, Albumin, and D-Dimers. The best ANN based on these indices achieved accuracy 95.97%, precision 90.63%, sensitivity 93.55%. and F1-score 92.06%, verified in the validation cohort. The preliminary findings reveal for the first time an ANN to predict ICU hospitalization accurately and early, using only 5 easily accessible laboratory indices.
The findings of the research are particularly important, because they can contribute to the better evaluation of patients who will need special medical care and treatment. Additionally, the proposed methodology can be applied to the evaluation of hematological parameters that can predict the outcome of patients with other serious diseases such as various forms of cancer. Based on the highly encouraging first results, an update of the database using hematological data from more than 1500 patients is underway, with the aim of both further documenting/updating the study and validating the reliability of the prediction.