Tools
Tools
Huang, Hui (2024) Enhancing the Efficiency of Heart Disease Prediction Using Cloud Machine Learning Techniques. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (770kB) | Preview |
Abstract
Heart disease has been a critical focus of international medical attention as it remains to be the major cause of global death. Accurate heart disease prediction is a useful part of medical solutions. Traditional prediction approaches often face efficiency limitations in terms of prolonged training time, suboptimal resource utilization, and difficulties in processing large-scale datasets. To address these challenges, this study investigated the potential advantages of using cloud machine learning techniques to enhance the efficiency and performance of heart disease prediction.
Logistic Regression, Random Forest, and XGBoost algorithms were implemented with automated hyperparameter tuning to optimize models in both environments. Random Forest algorithm was used to assess the impact of parallel processing on various nodes and processes. The results show significant improvements in distributed computing and automated hyperparameter tuning scenarios. The training time and resource utilization have been reduced by 75.2% and 18.9 % in Random Forest training with the configuration of 2 nodes and 2 processes for each node, while the accuracy and ROC AUC score increased by 0.30% and 0.25%. In the setup of hyperparameter optimization in a cloud-based environment, the training time in Logistic Regression and XGBoost has been reduced by 55.4% and 8.7 compared with a single-node local environment.
This study offers a practical solution for healthcare institutions to employ cloud computing for medical decisions or their clinical applications, especially those with restricted computational resources. It aims to narrow the gap between limited computational resources and the increasing demand for big data in heart disease instances.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Makki, Ahmed UNSPECIFIED |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing R Medicine > Healthcare Industry Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning |
| Divisions: | School of Computing > Master of Science in Cloud Computing |
| Depositing User: | Ciara O'Brien |
| Date Deposited: | 15 Jul 2025 11:23 |
| Last Modified: | 15 Jul 2025 11:23 |
| URI: | https://norma.ncirl.ie/id/eprint/8104 |
Actions (login required)
 |
View Item |