Related Experiment Video
Updated: Apr 30, 2026
06:57
Author Spotlight: Advancing Cardiovascular Imaging - Introducing the Spatially Weighted Calcium Score for Early Disease Detection
Published on: September 22, 2023
1.4K
Improved Inception-Capsule deep learning model with enhanced feature selection for early prediction of heart disease
Meghavathu S S Nayak1, Hussain Syed2
1School of Computer Science and Engineering (SCOPE), VIT-AP University, Amaravati, Andhra Pradesh, 522237, India.
Scientific Reports
|September 25, 2025
Summary
This study introduces IDLHICNet, a novel deep learning framework for accurate heart disease prediction. It effectively addresses challenges in medical datasets, improving early detection and patient outcomes.
Area of Science:
- Cardiology and Medical Informatics
- Artificial Intelligence in Healthcare
- Machine Learning for Disease Prediction
Background:
- Heart disease is a leading global cause of mortality, necessitating advanced prediction methods.
- Traditional machine learning models struggle with high-dimensional, imbalanced medical data.
- Existing prediction techniques often lack the precision required for early and effective intervention.
Purpose of the Study:
- To develop an advanced deep learning framework for accurate and early heart disease prediction.
- To enhance feature selection and data preprocessing for improved model performance.
- To overcome the limitations of traditional machine learning in handling complex medical datasets.
Main Methods:
- Proposed an Improved Deep Learning-based Hybrid Inception-Capsule Network (IDLHICNet) integrated with an Enhanced Whale Optimization Algorithm (EWOA) for feature selection.
- Implemented Improved K-Means Clustering (IKC) for outlier removal and SMOTE for class balancing.
- Utilized Min-Max normalization for feature scaling and combined Inception architecture with Capsule Networks for classification.
Main Results:
- Achieved high accuracy rates of 99.51%, 98.76%, and 99.07% on Faisalabad, CVD, and heart failure datasets, respectively.
- Demonstrated superior performance over state-of-the-art methods with enhanced precision, recall, and F1-scores.
- Validated the framework's effectiveness on multiple benchmark datasets for heart disease prediction.
Conclusions:
- The hybrid deep learning architecture (IDLHICNet) combined with sophisticated feature selection (EWOA) offers a powerful solution for heart disease prediction.
- The proposed methodology enables accurate and early detection, facilitating timely medical intervention and improved patient outcomes.
- This research highlights the potential of advanced AI techniques in addressing critical challenges in cardiovascular disease management.
