Hybrid Bayesian Network-Based Modeling: COVID-19-Pneumonia Case.

The primary goal of this paper is to develop an approach for predicting important clinical indicators, which can be used to improve treatment. Using mathematical predictive modeling algorithms, we examined the course of COVID-19-based pneumonia (CP) with inpatient treatment. Algorithms used include dynamic and ordinary Bayesian networks (OBN and DBN), popular ML algorithms, the state-of-the-art auto ML approach and our new hybrid method based on DBN and auto ML approaches. Predictive targets include treatment outcomes, length of stay, dynamics of disease severity indicators, and facts of prescribed drugs for different time intervals of observation. Models are validated using expert knowledge, current clinical recommendations, preceding research and classic predictive metrics. The characteristics of the best models are as follows: MAE of 3.6 days of predicting LOS (DBN plus FEDOT auto ML framework), 0.87 accuracy of predicting treatment outcome (OBN); 0.98 F1 score for predicting facts of prescribed drug (DBN). Moreover, the advantage of the proposed approach is Bayesian network-based interpretability, which is very important in the medical field. After the validation of other CP datasets for other hospitals, the proposed models can be used as part of the decision support systems for improving COVID-19-based pneumonia treatment. Another important finding is the significant differences between COVID-19 and non-COVID-19 pneumonia.

Predictive Modeling of COVID and non-COVID Pneumonia Trajectories.

Today pneumonia is one of the main problems of all countries around the world. This disease can lead to early disability, serious complications, and severe cases of high probabilities of lethal outcomes. A big part of cases of pneumonia are complications of COVID-19 disease. This type of pneumonia differs from ordinary pneumonia in symptoms, clinical course, and severity of complications. For optimal treatment of disease, humans need to study specific features of providing 19 pneumonia in comparison with well-studied ordinary pneumonia. In this article, the authors propose a new approach to identifying these specific features. This method is based on creating dynamic disease models for COVID and non-COVID pneumonia based on Bayesian Network design and Hidden Markov Model architecture and their comparison. We build models using real hospital data. We created a model for automatically identifying the type of pneumonia (COVID-19 or ordinary pneumonia) without special COVID tests. And we created dynamic models for simulation future development of both types of pneumonia. All created models showed high quality. Therefore, they can be used as part of decision support systems for medical specialists who work with pneumonia patients.