Predicting prognosis in COVID-19 patients using machine learning and readily available clinical data
Thomas W Campbell1, Melissa P Wilson2, Heinrich Roder1
1Biodesix, United States.
International Journal of Medical Informatics
|October 3, 2021
Summary
Machine learning models can predict severe COVID-19 outcomes in hospitalized patients using admission data. These tools help identify high-risk individuals for better treatment and improved patient outcomes.
Area of Science:
- Medical informatics
- Clinical prediction models
- Infectious disease epidemiology
Background:
- Hospitalized COVID-19 patients require prognostic tools to identify severe disease risk.
- Accurate risk stratification can guide treatment decisions and improve patient outcomes.
Purpose of the Study:
- To develop and validate machine learning models for stratifying COVID-19 patients by risk of severe outcomes.
- Utilize readily available hospital admission data for risk assessment.
Main Methods:
- Hierarchical ensemble classification models were trained on 229 hospitalized COVID-19 patients.
- Predictors included patient characteristics, vital signs, and basic lab results.
- Independent validation was performed on a separate cohort of 330 patients.
Main Results:
- Models achieved high precision (near or above 0.9) in identifying low-risk groups.
- A significant increase in severe outcomes was observed across increasing risk strata.
- Key predictors included C-reactive protein, lactate dehydrogenase, and D-dimer.
Conclusions:
- Machine learning models using routine admission data can effectively assess severe outcomes risk in COVID-19 patients.
- These prognostic tools aid in clinical decision-making for hospitalized individuals.
Related Concept Videos
01:18Steps in Outbreak Investigation
257
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
257
01:21Cancer Survival Analysis
481
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
481
01:11Residuals and Least-Squares Property
8.1K
The vertical distance between the actual value of y and the estimated value of y. In other words, it measures the vertical distance between the actual data point and the predicted point on the line
If the observed data point lies above the line, the residual is positive, and the line underestimates the actual data value for y. If the observed data point lies below the line, the residual is negative, and the line overestimates the actual data value for y.
The process of fitting the best-fit...
If the observed data point lies above the line, the residual is positive, and the line underestimates the actual data value for y. If the observed data point lies below the line, the residual is negative, and the line overestimates the actual data value for y.
The process of fitting the best-fit...
8.1K
01:03Prediction Intervals
2.5K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
2.5K