Cardiovascular Risk Prediction in Indian Youth Using Hybrid CNN–LSTM Approach with Feature Augmentation

Read full text Summarise with VIRAT
Authors: Shital Solanki, Dipesh Makwana
Cited by: 0
References: 10

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Cardiovascular disease is no longer confined to older populations is increasingly observed among young adults due to changing dietary habits, reduced physical activity, and persistent psychosocial stress.Conventional statistical and machine learning models often underperform in capturing nonlinear and sequential dependencies inherent in health and lifestyle data.In this study, National Family Health Survey-5 (NFHS-5) dataset used to build hybrid deep learning models that combine convolutional neural networks (CNNs) and long shortterm memory networks (LSTMs).In addition to standard features such as BMI, blood pressure, and anemia status, Feature augmentation techniques were applied to generate composite nutritional and lifestyle features.Comparative experiments demonstrated that the CNN-LSTM hybrid approach achieved superior performance with accuracy 92.1%, AUROC 0.91 compared to Logistic Regression, Random Forest, XGBoost, and standalone CNN or LSTM models.These results highlight the potential of hybrid deep learning for youth-focused preventive health in India.

Fields of Study & Subjects

Artificial Intelligence in Healthcare Cardiovascular Health and Risk Factors Health Promotion and Cardiovascular Prevention

Publication details

Journal / source
Advances in engineering research/Advances in Engineering Research
Pages
28–38
Publication date
30 March 2026
Language
EN
Document type
Book chapter