article · Neural Computing and Applications
Cardiovascular diseases present a substantial global health burden, creating a strong requirement for earlier detection and intervention. This research examines the use of convolutional neural networks to predict heart disease risk from personal health markers, including blood pressure, cholesterol levels, and lifestyle factors. The deep learning architecture combines embedding layers to convert categorical information into numerical data, convolutional layers to capture spatial patterns, and dense layers to assess complex interactions. Regularisation methods, such as dropout and batch normalisation, alongside hyperparameter tuning, were used to improve generalisation. When evaluated against traditional predictive methods, the network showed superior performance, achieving an R-squared value of 0.994. The work also highlights the necessity of model interpretability and ethical considerations to support responsible use in healthcare settings.
Cardiovascular conditions remain a leading cause of illness worldwide. Developing computational models that can reliably analyse complex mixtures of clinical indicators, such as blood pressure and cholesterol alongside lifestyle factors, helps identify individuals at elevated risk earlier. High predictive accuracy provides clinicians with better diagnostic support, assisting timely preventive care and personalised medical management.
The research points towards software-based clinical decision support tools for healthcare providers to aid cardiovascular risk management and prevention. Because the architecture relies on standard health metrics, it could be incorporated into hospital or diagnostic software. However, the abstract describes an analytical model validated against conventional methods rather than clinical deployment, indicating that the technology remains at an early, algorithmic stage requiring clinical integration and interpretability frameworks.
AI-generated from the published abstract. Always read the original work before citing.
Abstract Cardiovascular diseases (CVDs) remain a global burden, highlighting the need for innovative approaches for early detection and intervention. This study investigates the potential of deep learning, specifically convolutional neural networks (CNNs), to improve the prediction of heart disease risk using key personal health markers. Our approach revolutionizes traditional healthcare predictive modeling by integrating CNNs, which excel at uncovering subtle patterns and hidden interactions among various health indicators such as blood pressure, cholesterol levels, and lifestyle factors. To achieve this, we leverage advanced neural network architectures. The model utilizes embedding layers to transform categorical data into numerical representations, convolutional layers to extract spatial features, and dense layers to model complex interactions and predict CVD risk. Regularization techniques like dropout and batch normalization, along with hyperparameter optimization, enhance model generalizability and performance. Rigorous validation against conventional methods demonstrates the model’s superiority, with a significantly higher R 2 value of 0.994. This achievement underscores the model’s potential as a valuable tool for clinicians in CVD prevention and management. The study also emphasizes the need for interpretability in deep learning models and addresses ethical considerations to ensure responsible implementation in clinical practice.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1007/s00521-024-10453-2
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
