Related Experiment Video
Updated: Jun 8, 2026
Hydra, a Computer-Based Platform for Aiding Clinicians in Cardiovascular Analysis and Diagnosis
Published on: September 26, 2018
Cardiovascular Disease Risk Stratification Using Hybrid Deep Learning Paradigm: First of Its Kind on Canadian Trial
Mrinalini Bhagawati1, Sudip Paul1, Laura Mantella2
1Department of Biomedical Engineering, North-Eastern Hill University, Shillong 793022, India.
Insights
Hybrid deep learning (HDL) models significantly improve cardiovascular disease (CVD) risk prediction from carotid plaques compared to traditional machine learning. AtheroEdge™ 3.0HDL demonstrates superior performance and clinical adaptability for enhanced CVD risk stratification.
Area of Science:
- Artificial Intelligence in Medical Imaging
- Cardiovascular Disease Risk Prediction
- Deep Learning for Healthcare
Background:
- Cardiovascular disease (CVD) risk assessment traditionally relies on carotid plaque evaluation.
- AtheroEdge™ 3.0HDL was developed to leverage carotid plaque features for improved CVD risk prediction.
- The study hypothesizes that hybrid deep learning (HDL) surpasses other deep learning and machine learning (ML) approaches.
Purpose of the Study:
- To evaluate the efficacy of AtheroEdge™ 3.0HDL in predicting cardiovascular disease (CVD) risk.
- To compare the performance of hybrid deep learning (HDL) models against traditional machine learning (ML) and other deep learning paradigms.
- To validate the reliability and clinical adaptability of the AtheroEdge™ 3.0HDL system.
Main Methods:
- A study cohort of 500 individuals with carotid B-mode ultrasonography and coronary angiography data was analyzed.
- Machine learning (ML) feature selection utilized principal component analysis (PCA), chi-square test (CST), and random forest regression (RFR).
- Six novel hybrid deep learning (HDL) models were developed and validated against unidirectional/bidirectional deep learning and ML models using seen and unseen datasets.
Main Results:
- The hybrid deep learning (HDL) system demonstrated a 30.20% performance improvement over the machine learning (ML) system on seen datasets (0.954 vs. 0.702).
- ML feature extraction analysis revealed 70% commonality across PCA, CST, and RFR methods.
- AtheroEdge™ 3.0HDL exhibited excellent generalization, with less than 1% difference between seen and unseen data (p-value < 0.001), meeting regulatory standards.
Conclusions:
- The hypothesis was scientifically validated, confirming AtheroEdge™ 3.0HDL's superior performance in CVD risk prediction.
- The AtheroEdge™ 3.0HDL model demonstrated reliability and stability through rigorous testing.
- The system is clinically adaptable, offering potential for improved cardiovascular disease risk stratification.
Background:
The risk of cardiovascular disease (CVD) has traditionally been predicted via the assessment of carotid plaques. In the proposed study, AtheroEdge™ 3.0HDL (AtheroPoint™, Roseville, CA, USA) was designed to demonstrate how well the features obtained from carotid plaques determine the risk of CVD. We hypothesize that hybrid deep learning (HDL) will outperform unidirectional deep learning, bidirectional deep learning, and machine learning (ML) paradigms.
Methodology:
500 people who had undergone targeted carotid B-mode ultrasonography and coronary angiography were included in the proposed study. ML feature selection was carried out using three different methods, namely principal component analysis (PCA) pooling, the chi-square test (CST), and the random forest regression (RFR) test. The unidirectional and bidirectional deep learning models were trained, and then six types of novel HDL-based models were designed for CVD risk stratification. The AtheroEdge™ 3.0HDL was scientifically validated using seen and unseen datasets while the reliability and statistical tests were conducted using CST along with p-value significance. The performance of AtheroEdge™ 3.0HDL was evaluated by measuring the p-value and area-under-the-curve for both seen and unseen data.
Results:
The HDL system showed an improvement of 30.20% (0.954 vs. 0.702) over the ML system using the seen datasets. The ML feature extraction analysis showed 70% of common features among all three methods. The generalization of AtheroEdge™ 3.0HDL showed less than 1% (p-value < 0.001) difference between seen and unseen data, complying with regulatory standards.
Conclusions:
The hypothesis for AtheroEdge™ 3.0HDL was scientifically validated, and the model was tested for reliability and stability and is further adaptable clinically.
