Explainable machine learning for real-time deterioration alert prediction to guide pre-emptive treatment
Aida Brankovic1, Hamed Hassanzadeh2, Norm Good2
1CSIRO Australian e-Health Research Centre, Brisbane, QLD, 4029, Australia. aida.brankovic@csiro.au.
Scientific Reports
|July 11, 2022
Summary
This study introduces a machine learning tool using Electronic Medical Record (EMR) data to predict patient deterioration. It helps clinicians prioritize care by identifying high-risk individuals early, improving patient outcomes.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Healthcare
- Clinical Decision Support
Background:
- Electronic Medical Records (EMR) offer potential for efficient patient care management.
- Clinical decision support tools are crucial for identifying adverse events and acute illnesses.
- Early identification of patient deterioration is vital for timely intervention.
Purpose of the Study:
- To develop and evaluate a machine learning-driven tool for predicting patients at high risk of critical conditions.
- To provide a pre-emptive solution for clinicians to prioritize patient evaluation based on deterioration risk.
- To offer visualized explanations of contributing factors for clinical decision-making.
Main Methods:
- Development of a machine learning tool utilizing real-time Electronic Medical Record (EMR) data.
- Application of the tool to a test cohort of 18,648 patient records.
- Evaluation of prediction accuracy and lead time for patient deterioration.
Main Results:
- The tool achieved 100% sensitivity in predicting patient deterioration.
- Predictions were accurate for windows 2-8 hours in advance.
- The tool identified patients at 95%, 85%, and 70% risk of deterioration.
Conclusions:
- Machine learning tools analyzing EMR data can effectively predict patient deterioration.
- Early risk identification enables proactive clinical intervention and resource prioritization.
- Visualized explanations enhance the tool's utility for guiding clinical decisions.
Related Concept Videos
01:18Steps in Outbreak Investigation
182
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:
182
01:19Survival Tree
153
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a...
Building a Survival Tree
Constructing a...
153
01:03Prediction Intervals
2.3K
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.
2.3K