This technology defines a set of ECG features for predicting the onset of cardiac arrest. In-Hospital Cardiac Arrest (I-HCA) effects some 200,000 people yearly, and has a survival rate under 30%. Despite the fact that some ECG patterns are known to be associated with cardiac arrest, there are no reliable tools to predict its onset. This technology predicts cardiac arrests before they occur with high sensitivity and specificity. This degree of accuracy creates a tool which will allow doctors to respond confidently and quickly to oncoming cardiac arrest, greatly improving patient survivability.
Data-Driven Prediction of Cardiac Arrest
The proposed technology monitors ECG signals and uses a defined set of predictors to detect oncoming cardiac arrest. These predictors were developed though a large data survey of patients who had suffered I-HCA. The resulting set of predictors detects 77.6% of oncoming cardiac arrest cases, as determined through analyzing prior cardiac arrest data. This technology will reduce the 150000 deaths resulting from I-HCA annually.
- Monitoring—in and out of hospital—of patients with a history of cardiac arrest to predict future cardiac complications. *Improving survival rates of patients suffering cardiac arrest.
- Higher degree of sensitivity and specificity than any other existing method.
- Uses a single data type and can monitor patients in real-time.