Abstract
Cardiac arrest occurs more often in younger people. Those who are dying at a younger age these days due to health instability should seek immediate medical attention. When the heart breaks down and abruptly stops beating, it can cause an irregular pulse, which can lead to cardiac arrest. Rapid loss of awareness, cessation of respiration, or scarcely audible gasps occur. Death happens in a few minutes. There are a tonne of cardiac arrest datasets available in the healthcare sector. By collecting real-time data from sensors and feeding it into a prediction model, it is possible to detect cardiac arrests with a low-to-high probability. To make model predictions, machine learning algorithms are being used. In this present circumstance, we employ six particular machine learning algorithms: decision tree, logistic regression, random forest classifier, support vector machine, and K-nearest neighbour. The most effective of these methods is the decision tree and random forest classification. The model's forecast accuracy is 100 percent. Given that it has a number of advantages over the other, the random forest classifier is the best among the two finest algorithms. Finally, the proposed model is used to predict whether or not a cardiac arrest will occur within the next ten years, while also being used outside of the hospital. This will help alert the patient to consult the doctor.
Keyword
Cardiac arrest, machine learning, smart watches, sphygmomanometer, and glucometer.
PDF Download (click here)
|