Identifying COVID-19 diseases in Heart Patients using Cardio IoT algorithms

Abstract

The COVID-19 pandemic has posed significant health challenges, particularly for individuals with underlying cardiovascular conditions, who are at a higher risk of severe outcomes. Early detection and accurate risk assessment are crucial to mitigating complications and improving patient outcomes. Leveraging Internet of Things (IoT) technology, this research introduces a robust framework for the prediction and validation of COVID-19 cases among cardiovascular patients. The proposed framework is built on the Deep Boosted Cardiovascular Risk Algorithm (DBCRA), a novel machine learning approach that integrates IoT-driven real-time data collection with advanced decision-making capabilities. By analyzing diverse physiological parameters captured through IoT devices, the DBCRA identifies at-risk individuals with enhanced precision.

The research highlights the effectiveness of the DBCRA in terms of prediction accuracy, reliability, and computational efficiency, surpassing traditional models in large-scale and dynamic IoT environments. The experimental analysis demonstrates the algorithm's ability to handle multi-dimensional IoT data streams while maintaining scalability and robustness. Metrics such as the Adjusted Rand Index, Silhouette Score, and Dice Similarity Coefficient validate the performance, reflecting significant improvements in prediction outcomes. The proposed approach provides a transformative solution for integrating IoT and artificial intelligence in healthcare, enabling personalized risk management and timely interventions for cardiovascular patients impacted by COVID-19.