AI Models Predict COVID-19 Severity and Treatment for Hospitalized Patients

AI-Driven Decision Support System for COVID-19 Patients

Researchers from Florida Atlantic University's Christine E. Lynn College of Nursing and College of Engineering and Computer Science, in collaboration with Memorial Healthcare System, have developed an innovative AI-driven decision support system to predict the severity of COVID-19 and determine the best therapeutic interventions for hospitalized patients 1

1

2

Study Design and Methodology

The study analyzed electronic health record (eHR) data from 5,371 patients admitted to a South Florida hospital with COVID-19 between March 2020 and January 2021. Three Random Forest models were trained to predict the need for:

1. Intensive Care Unit (ICU) admission with mechanical ventilation
2. ICU admission without mechanical ventilation
3. Intermediate Care Unit (IMCU) admission

The models utilized 24 variables, including sociodemographics, comorbidities, and medications, focusing on data collected at the time of hospital admission 1

1

2

Key Findings and Predictors

The study, published in the journal Diagnostics in early fall 2024, identified several critical factors influencing COVID-19 severity:

• Age, race, sex, and body mass index (BMI)
• Pre-existing conditions: diabetes, hypertension, early stages of kidney disease, and pneumonia
• Presence of diarrhea

Individuals aged 65 and older, males, current smokers, and those classified as overweight or obese were found to be at greater risk of severe illness 1

1

2

Interaction of Risk Factors

The research explored the co-occurrence of risk factors, revealing important interactions:

• Pneumonia combined with diabetes increased the risk of mechanical ventilation
• Diarrhea interacted strongly with diabetes for ICU admissions
• IMCU severity was linked to combinations of diarrhea with pneumonia and hypertension in older adults 1

  1

  2

  2

Related Stories

AI Revolutionizes Long COVID Care: Identifying Patient Subgroups and Tailoring Hospital Resources

AI Model Revolutionizes Sepsis Mortality Prediction in ICUs

AI-enabled tool cuts hospital deaths 18% by detecting patient deterioration earlier

AI Tool Predicts Underfeeding Risk in ICU Patients on Ventilators, Enabling Early Intervention

Medication Effects and Model Interpretability

The study found that medications such as angiotensin II receptor blockers and ACE inhibitors appeared to lower disease severity, aligning with prior research on their protective effects. The top features identified by the models' interpretability were from the "sociodemographic characteristics," "pre-hospital comorbidities," and "medications" categories 1

1

2

Implications and Future Applications

This novel approach stands out by using readily accessible eHR data and combining machine learning interpretability techniques with traditional statistical methods. The findings provide actionable insights for improving patient care and supporting healthcare systems during high-demand conditions 1

1

2

Dr. Debarshi Datta, senior author and assistant professor at FAU's Christine E. Lynn College of Nursing, emphasized the broader implications: "Developing an AI-driven decision support system to predict critical clinical events in COVID-19 in-patients not only meets the urgent demands of a pandemic but also breaks new ground in AI and machine learning in healthcare" 1

1

2

The application of AI/machine learning in healthcare extends beyond COVID-19, holding promise for improving diagnosis, treatment selection, disease surveillance, and patient outcomes across various medical specialties and healthcare settings. This knowledge empowers public health authorities to proactively plan and implement targeted interventions, potentially mitigating the impact of future disease outbreaks and optimizing healthcare delivery 1

1

2