Smart Cardiac Framework for an Early Detection of Cardiac Arrest Condition and Risk

Area of Science:

• Cardiology and Health Informatics
• Machine Learning Applications in Healthcare

Background:

• Cardiovascular disease (CVD) poses a significant global health challenge, with accurate prediction of events like cardiac arrest remaining difficult.
• Existing healthcare datasets may yield erroneous predictions, necessitating the use of real-time data for improved accuracy.

Purpose of the Study:

• To develop an accurate and economical system for predicting cardiovascular disease risk using real-time data.
• To introduce novel gender-based and age-wise risk classification approaches for survival probability assessment.

Main Methods:

• Collected real-time data using sensors and stored it on Google Firebase.
• Classified data using six machine learning algorithms: Artificial Neural Network (ANN), Random Forest Classifier (RFC), Gradient Boost Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Naïve Bayes (NB), and Decision Tree (DT).
• Employed Kaplan-Meier and Cox regression survival analysis for risk detection and classification, including gender- and age-specific models.

Main Results:

• The Decision Tree (DT) algorithm achieved an overall accuracy of 98% for cardiac risk prediction on the collected dataset.
• The novel gender-based and age-wise risk classification models accurately predicted survival probabilities.
• The proposed system offers a cost-effective alternative to existing prediction systems.

Conclusions:

• Real-time data collection and machine learning, particularly the Decision Tree algorithm, significantly enhance the accuracy of cardiac risk prediction.
• The developed gender- and age-specific models provide valuable tools for healthcare professionals to assess individual patient risk and survival probability.
• The proposed system presents an economical and effective solution for improving cardiovascular disease risk management.