Key result
SecureCardioNet achieves ~83% accuracy for five-class cardiovascular disease classification.
Why the study?
Cardiovascular diseases require intelligent and secure healthcare monitoring frameworks capable of supporting real-time, multi-class diagnosis in Wireless Body Area Network (WBAN) environments, moving beyond conventional binary heart disease prediction approaches.
Does the SecureCardioNet blockchain-enabled deep learning framework improve five-class cardiovascular disease classification accuracy and data security in Wireless Body Area Network (WBAN) environments?
Does the SecureCardioNet blockchain-enabled deep learning framework improve five-class cardiovascular disease classification accuracy and data security in Wireless Body Area Network (WBAN) environments?
SecureCardioNet demonstrates 82.5% accuracy and 0.283-second blockchain latency for secure, real-time five-class cardiovascular disease classification using WBAN sensor data.
Hypothesis-generating for blockchain-secured DL classification in WBANs; prospective validation required before clinical consideration.
Cardiovascular diseases remain one of the leading causes of mortality worldwide, requiring intelligent and secure healthcare monitoring frameworks capable of supporting real-time diagnosis in Wireless Body Area Network (WBAN) environments.This study proposes SecureCardioNet, a blockchain-enabled deep learning framework for five-class cardiovascular disease classification using optimized physiological feature sequences derived from WBAN sensors.Unlike conventional binary heart disease prediction approaches, the proposed framework performs multi-class assessment for Arrhythmia, Diabetes, Healthy, Hypertension, and Hypotension conditions.The preprocessing pipeline incorporates Interquartile Range (IQR)-based outlier removal, Z-score normalization using StandardScaler, Synthetic Minority Over-sampling Technique (SMOTE) balancing, and ANOVA F-score feature selection to construct a clinically meaningful 30-dimensional feature vector from ECG morphology, vital signs, and patient metadata.SecureCardioNet integrates multi-scale 1D Convolutional Neural Networks with kernel sizes of 1, 3, and 5 together with a dual-attention mechanism consisting of channel attention and spatial attention modules to improve discriminative feature learning.To ensure secure and tamper-resistant medical data management, the framework incorporates an optimized iterative SHA-256 blockchain workflow within an Ethereum-enabled private consortium network.Experimental evaluation using 10-fold cross-validation demonstrates that SecureCardioNet achieves a mean classification accuracy of 82.5%, an F1-score of 0.825, and ROC-AUC values ranging from 0.89 to 0.96 across all disease categories.In addition, the optimized blockchain implementation achieves a low processing latency of 0.283 seconds, satisfying real-time WBAN monitoring requirements.The obtained results confirm the effectiveness of SecureCardioNet for secure, accurate, and scalable cardiovascular healthcare analytics.
No takes yet. Share an insight, caveat, or question.
Kayalvizhi et al. (2026) studied Cardiovascular disease (Arrhythmia, Diabetes, Healthy, Hypertension, and Hypotension). SecureCardioNet (blockchain-enabled multi-scale 1D-CNN with dual-attention framework) vs. Conventional machine learning and deep learning models was evaluated on Five-class cardiovascular disease classification accuracy. SecureCardioNet achieved a mean five-class cardiovascular disease classification accuracy of 82.5%, an F1-score of 0.825, and a blockchain latency of 0.283 seconds.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: