Abstract

Heart disease remains one of the leading global causes of mortality, necessitating accurate and interpretable computational models for early diagnosis. This study introduces a hybrid deep-learning framework named CBGWO, integrating convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) with grey wolf optimization (GWO), to enhance predictive accuracy on the Cleveland Heart Disease Dataset. The dataset includes 303 patient records with 13 clinically validated features, processed through median imputation, feature standardization, one-hot encoding, and SMOTE-based balancing to improve sensitivity toward minority disease cases. The CNN extracts spatial feature interactions, while the BiLSTM captures bidirectional temporal-like correlations among medical variables. GWO performs automated hyperparameter tuning and feature-weight assignment, optimizing filter size, learning rate, dropout, hidden units, and batch size through a medically motivated fitness function that prioritizes F1-score and Area Under Curve (AUC) while penalizing redundant features. Experimental evaluation demonstrates that the proposed CBGWO model significantly outperforms baseline architectures, achieving 93.50% accuracy, 93.10% precision, 93.40% recall, and 93.20% F1-score, surpassing standalone CNN, BiLSTM, and their GWO-integrated variants. ROC-curve and confusion-matrix analysis confirms improved discrimination capability and reduced false-negative risk, which is crucial in clinical diagnosis to avoid missed cardiac-risk cases. These outcomes highlight the value of combining deep-learning architectures with evolutionary optimization for precise and reliable medical decision-support systems.

Keywords

Heart disease prediction, Deep learning, Grey wolf optimization, CNN-BiLSTM, Cleveland Dataset.

Cite this article

Kumar S, Gautam S. CBGWO: a hybrid CNN-BiLSTM and grey wolf optimization framework for heart disease prediction. International Journal of Advanced Computer Research. 2026;16(74):1-9. DOI : 10.19101/IJACR.2025.1570029

References

[1] Karna VV, Karna VR, Janamala V, Devana VK, Ch VR, Tummala AB. A comprehensive review on heart disease risk prediction using machine learning and deep learning algorithms. Archives of Computational Methods in Engineering. 2025; 32(3):1763-95.

[2] Kumar A, Dhanka S, Sharma A, Bansal R, Fahlevi M, Rabby F, et al. A hybrid framework for heart disease prediction using classical and quantum-inspired machine learning techniques. Scientific Reports. 2025; 15(1):1-31.

[3] Rohan D, Reddy GP, Kumar YP, Prakash KP, Reddy CP. An extensive experimental analysis for heart disease prediction using artificial intelligence techniques. Scientific Reports. 2025; 15(1):1-47.

[4] Mulani AO, Sardey MP, Kinage K, Salunkhe SS, Fegade T, Fegade PG. ML-powered internet of medical things (MLIOMT) structure for heart disease prediction. Journal of Pharmacology and Pharmacotherapeutics. 2025; 16(1):38-45.

[5] Kushwaha PK, Dagur A, Shukla D. HSLE: a hybrid ensemble classifier for prediction of heart disease. Recent Advances in Electrical & Electronic Engineering. 2025; 18(7):781-92.

[6] Talaat FM, Aly WF. Toward precision cardiology: a transformer-based system for adaptive prediction of heart disease. Neural Computing and Applications. 2025; 37(19):13547-71.

[7] Kumar PR, Chakrabarti P, Chakrabarti T, Unhelkar B, Margala M. Heart disease prediction using spark architecture with fused feature set and hybrid squeezenet-linknet model. Biomedical Signal Processing and Control. 2025; 100:107070.

[8] Sivasubramaniam S, Balamurugan SP. Early detection and prediction of heart disease using wearable devices and deep learning algorithms. Multimedia Tools and Applications. 2025; 84(9):6187-201.

[9] Talaat FM. Revolutionizing cardiovascular health: integrating deep learning techniques for predictive analysis of personal key indicators in heart disease. Neural Computing and Applications. 2025; 37(1):1-24.

[10] Balamurugan M, Meera DS. Hybrid optimized temporal convolutional networks with long short-term memory for heart disease prediction with deep features. Computer Methods in Biomechanics and Biomedical Engineering. 2025; 28(7):996-1020.

[11] Aabdalla IDS, Vasumathi D. A novel hybrid deep learning and reinforcement learning framework for multimodal cardiovascular disease prediction. International Journal of Advanced Computer Research. 2025;15(73):25-38.

[12] Mostofi S, Kordrostami S, Refahi SAH, Faridi MM, Shokri S. Diagnosis of heart disease using an advanced triple hybrid algorithm combining machine learning techniques. Journal of Modelling in Management. 2025; 20(2):668-700.

[13] Kaur J, Khehra BS, Singh A. A state of the art survey of soft computing based approaches for heart disease detection and development of a novel PSO-fuzzy approach. Archives of Computational Methods in Engineering. 2025:1-24.

[14] Narasimhan G, Victor A. A hybrid approach with metaheuristic optimization and random forest in improving heart disease prediction. Scientific Reports. 2025; 15(1):1-18.

[15] Ashwini A, Chirchi V, Balasubramaniam S, Shah MA. Bio inspired optimization techniques for disease detection in deep learning systems. Scientific Reports. 2025; 15(1):1-30.

[16] Begum F, Valan JA. Enhancing heart disease diagnosis with meta-heuristic algorithms: a combined HHO and PSO approach. Biomedical Materials & Devices. 2026; 4(1):574-89.

[17] Roopa T, Ganesh DR. GPCB: a hybrid GA-PSO and Transformer-guided CNN-BiLSTM framework for cardiovascular disease prediction. Ingenierie des Systemes d'Information. 2025; 30(8):1985-99.

[18] Gupta I, Bajaj A, Malhotra M, Sharma V, Abraham A. Heart disease prediction using a hybrid feature selection and ensemble learning approach. IEEE Access. 2025; 13: 111926-37.

[19] Jia Z, Zhou Z. Heart disease prediction model based on GBDT-RF. In 4th international symposium on computer applications and information technology (ISCAIT) 2025 (pp. 302-6). IEEE.

[20] Manocha M, Kumar A, Singh VK, Singh Y. Heart disease prediction analysis with machine learning. In 3rd international conference on disruptive technologies (ICDT) 2025 (pp. 1378-82). IEEE.

[21] Shinde PV, Sanghavi MR. Enhanced and effective model for heart disease prediction using hybridization of random forest and adaboost (HRFAB). In international conference on machine learning and autonomous systems (ICMLAS) 2025 (pp. 674-9). IEEE.

[22] Sharma G, Upadhyay A, Chaudhary V, Kulshrestha D. IoT-enabled heart disease risk prediction using machine learning algorithms: a comparative analysis of ML models. In IEEE Guwahati subsection conference (GCON) 2025 (pp. 1-5). IEEE.

[23] Zhong J, Wei D, Zhuo Y, Yin M, Kang Y, Shi W. Feasibility study and practice of machine learning-based heart disease prediction. In 8th international conference on electronics technology (ICET) 2025 (pp. 492-6). IEEE.

[24] Rao T, Kodipalli A. Heart disease prediction using AHP and fuzzy TOPSIS methods. In international conference on cognitive computing in engineering, communications, sciences and biomedical health informatics (IC3ECSBHI) 2025 (pp. 1029-33). IEEE.

[25] Dwivedi AK, Yadav A, Gamad RS. Heart disease prediction using hybrid machine learning with ECG dataset for healthcare-IOT application. In 4th OPJU international technology conference (OTCON) on smart computing for innovation and advancement in industry 5.0 2025 (pp. 1-6). IEEE.

[26] Shah N, Poonia A. Heart disease prediction system using ensemble machine learning technique. In 5th international conference on soft computing for security applications (ICSCSA) 2025 (pp. 945-9). IEEE.

[27] Liu S. Heart disease prediction using optimized machine learning models. In 6th international conference on big data & artificial intelligence & software engineering (ICBASE) 2025 (pp. 464-70). IEEE.