Predicting the Evolution of COVID-19 Mortality Risk: a Recurrent Neural Network Approach

Villegas, M.; Gonzalez-Agirre, A.; Gutierrez-Fandino, A.; Armengol-Estape, J.; Carrino, C. P.; Perez Fernandez, D.; Soares, F.; Serrano, P.; Pedrera, M.; Garcia, N.; Valencia, A.

Background: The propagation of COVID-19 in Spain prompted the declaration of the state of alarm on March 14, 2020. On 2 December 2020, the infection had been confirmed in 1,665,775 patients and caused 45,784 deaths. This unprecedented health crisis challenged the ingenuity of all professionals involved. Decision support systems in clinical care and health services management were identified as crucial in the fight against the pandemic. Methods: This study applies Deep Learning techniques for mortality prediction in COVID-19 patients. Two datasets with clinical information (medication, laboratory tests, vital signs etc.) of 2,307 and 3,870 COVID-19 infected patients admitted to two Spanish hospital chains were used. Firstly, we built a sequence of temporal events gathering all the clinical information for each patient. Next, we used the temporal sequences to train a Recurrent Neural Network (RNN) model with an attention mechanism exploring interpretability. We conducted extensive experiments and trained the RNNs in different settings, performing hyperparameter search and cross-validation. We ensembled resulting RNNs to reduce variability and enhance sensitivity. Results: We assessed the performance of our models using both global metrics, by averaging the performance across all the days in the sequences. We also measured day-by-day metrics starting from the day of hospital admission and the outcome day and evaluated the daily predictions. Regarding sensitivity, when compared to more traditional models, our best two RNN ensemble models outperform a Support Vector Classifier in 6 and 16 percentage points, and Random Forest in 23 and 18 points. For the day-by-day predictions from the outcome date, the models also achieved better results than baselines showing system's ability towards early predictions. Conclusions: We have shown the feasibility of our approach to predict the clinical outcome (i.e. discharged alive or death) of patients infected with SARS-CoV-2. The result is a time series model that can support decision-making in healthcare systems and aims at interpretability. Despite the low-resource scenario, the results achieved are promising and suggests that more data will further increase the performance of the model.