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

Abstract

Cardiovascular disease (CVD) is considered to be one of the most epidemic diseases in the world today. Predicting CVDs, such as cardiac arrest, is a difficult task in the area of healthcare. The healthcare industry has a vast collection of datasets for analysis and prediction purposes. Somehow, the predictions made on these publicly available datasets may be erroneous. To make the prediction accurate, real-time data need to be collected. This study collected real-time data using sensors and stored it on a cloud computing platform, such as Google Firebase. The acquired data is then classified using six machine-learning algorithms: Artificial Neural Network (ANN), Random Forest Classifier (RFC), Gradient Boost Extreme Gradient Boosting (XGBoost) classifier, Support Vector Machine (SVM), Naïve Bayes (NB), and Decision Tree (DT). Furthermore, we have presented two novel gender-based risk classification and age-wise risk classification approach in the undertaken study. The presented approaches have used Kaplan-Meier and Cox regression survival analysis methodologies for risk detection and classification. The presented approaches also assist health experts in identifying the risk probability risk and the 10-year risk score prediction. The proposed system is an economical alternative to the existing system due to its low cost. The outcome obtained shows an enhanced level of performance with an overall accuracy of 98% using DT on our collected dataset for cardiac risk prediction. We also introduced two risk classification models for gender- and age-wise people to detect their survival probability. The outcome of the proposed model shows accurate probability in both classes.

Actionable absolute risk prediction of atherosclerotic cardiovascular disease based on the UK Biobank

“…The frequency of social visits could be indicative of someone's current mental health status, which has been shown to be a relevant CVD risk factor [58,59]. These and other non-laboratory risk factors could be collected by means of a questionnaire or passively deduced using data analytics from data sources such as GPS, calendar and sensors [26,60] from e.g. smartphones, smartwatches and fitness trackers.…”

“…Previous studies in this area use a limited set of risk factors and outcomes for their analyses [7,25,26]. In recent years, the knowledge of behavioral risk factors and of the pathophysiology of atherosclerotic CVDs have advanced tremendously [11,25].…”

A Web-Based Prediction Model for Cancer-Specific Survival of Elderly Patients With Clear Cell Renal Cell Carcinoma: A Population-Based Study

“…Awais et al ( 15 ) used texture analysis techniques in fluorescence imaging to help dentists identify areas of oncogenic cavity abnormalities in the oral cavity, and thus, perform biopsies more efficiently to predict the histological diagnosis of epithelial dysplasia. Shah et al ( 16 ) used machine learning to detect the condition and risk of cardiac arrest early to improve survival in patients with heart disease.…”

Analysis of Tracheobronchial Diverticula Based on Semantic Segmentation of CT Images via the Dual-Channel Attention Network

“…Ghazal et al reviewed the machine learning in smart healthcare ( 10 ) and an improved SVM method for diagnosing Hepatitis C ( 11 ). Shah et al ( 12 ) proposed two gender-based risk classification and age-based risk classification methods to identify cardiovascular diseases (CVD). The accuracy of the model can reach 98%, which improved the identification efficiency.…”