Blockchain-aided comparative study of heart disease detection using machine learning-based approaches with an...

Area of Science:

• Biomedical Engineering
• Computer Science
• Public Health

Background:

• Cardiovascular disease (CVD) is a leading cause of global mortality, accounting for 31% of all deaths annually.
• The critical need for advanced detection and secure data management in cardiovascular health is evident.
• Current challenges include accurate prediction and safeguarding sensitive patient information.

Purpose of the Study:

• To develop an integrated system for accurate cardiovascular disease detection.
• To enhance the security, transparency, and integrity of medical data using encryption and blockchain.
• To mitigate the impact of cardiovascular disease through improved diagnostic capabilities.

Main Methods:

• Integration of machine learning models (decision trees, random forests, Naïve Bayes, KNN, neural networks) for disease prediction.
• Implementation of a robust encryption algorithm for data security.
• Utilization of a private blockchain framework for tamper-proof storage and data integrity.

Main Results:

• A voting ensemble technique combined with various machine learning models achieved an accuracy of 89.2% in cardiovascular disease detection.
• The system successfully integrates predictive modeling with secure, decentralized data storage.
• Demonstrated enhanced accuracy and data security for cardiovascular health applications.

Conclusions:

• The proposed system offers a significant advancement in cardiovascular disease detection and management.
• The combination of machine learning and blockchain provides a secure and accurate platform for medical data.
• Implementation promises improved patient outcomes and more effective management of CVD-related challenges.