Prognosis of COVID-19 severity using DERGA, a novel machine learning algorithm
Panagiotis G Asteris1, Amir H Gandomi2, Danial J Armaghani3
1Computational Mechanics Laboratory, School of Pedagogical and Technological Education, Athens, Greece.
Insights
Predicting COVID-19 intensive care unit (ICU) admission risk is crucial. A new algorithm using four basic blood tests achieved 97.12% accuracy in identifying high-risk patients, aiding clinical management.
Area of Science:
- Medical informatics
- Hematology
- Infectious diseases
Background:
- Accurate prediction of COVID-19 intensive care unit (ICU) admission risk is vital for emergency department patient management.
- Existing models may require complex data, hindering rapid clinical application.
Purpose of the Study:
- To develop and validate a novel predictive model for ICU admission in COVID-19 patients.
- To identify key hematological indices for accurate risk stratification.
Main Methods:
- Utilized artificial neural networks and a Data Ensemble Refinement Greedy Algorithm (DERGA).
- Trained and tested the model on a database of 1596 COVID-19 patients.
- Evaluated model performance using 15 accessible hematological indices.
Main Results:
- Identified an optimal combination of four hematological indicators: neutrophil-to-lymphocyte ratio (NLR), lactate dehydrogenase, ferritin, and albumin.
- Achieved a high prediction accuracy of 97.12% for ICU admission risk.
- Demonstrated the model's reliability on a separate testing dataset.
Conclusions:
- The novel DERGA model provides a robust and accurate method for predicting COVID-19 ICU admission risk.
- The model relies solely on easily accessible basic hematological parameters.
- This approach can optimize clinical decision-making and patient management for COVID-19.
Abstract:
It is important to determine the risk for admission to the intensive care unit (ICU) in patients with COVID-19 presenting at the emergency department. Using artificial neural networks, we propose a new Data Ensemble Refinement Greedy Algorithm (DERGA) based on 15 easily accessible hematological indices. A database of 1596 patients with COVID-19 was used; it was divided into 1257 training datasets (80 % of the database) for training the algorithms and 339 testing datasets (20 % of the database) to check the reliability of the algorithms. The optimal combination of hematological indicators that gives the best prediction consists of only four hematological indicators as follows: neutrophil-to-lymphocyte ratio (NLR), lactate dehydrogenase, ferritin, and albumin. The best prediction corresponds to a particularly high accuracy of 97.12 %. In conclusion, our novel approach provides a robust model based only on basic hematological parameters for predicting the risk for ICU admission and optimize COVID-19 patient management in the clinical practice.
