Literature Survey on Early Disease Detection using Machine Learning

This paper presented MFF-HDNet, a multimodal fusion framework for early cardiovascular disease prediction that in-tegrates ECG temporal signals and structured clinical-lifestyle data through a dual-pathway deep learning architecture with cross-attention fusion. The proposed system addresses three fundamental limitations of prior work: single-modality data reliance, susceptibility to class imbalance, and lack of model interpretability. Experimental results on the UCI Cleveland and Framing-ham Heart Study datasets demonstrate classification accura-cies of 97.8% and 98.6%, AUC-ROC values of 0.978 and 0.984, and consistently competitive F1-scores, outperform-ing multiple state-of-the-art baseline methods. The ablation study confirms that each architectural component—the SE-FCN encoder, the 1D-CNN-GRU encoder, the cross-attention fusion module, and SMOTE-ENN resampling—contributes in-dependently and additively to the observed performance gains. SHAP-based analysis further validates clinically coherent fea-ture importance rankings, providing transparent explanations aligned with established cardiovascular risk factors. Looking ahead, future work will focus on: (a) integration of imaging modalities such as echocardiography and coro-nary CT angiography for richer multimodal representation; (b) federated learning protocols to enable privacy-preserving multi-institutional model training; (c) deployment within IoT-enabled wearable cardiovascular monitoring platforms for continuous real-time risk assessment; and (d) incorporation of genetic and genomic biomarkers to support personalised car-diac risk stratification for genetically predisposed individuals

[1]     P. Wagner, N. Strodthoff, R. Bousseljot, D. Kreiseler, A. I. Lunze, W. Samek, and T. Schaeffter, “PTB-XL, a Large Publicly Available Electrocardiography Dataset,” Scientific Data, vol. 7, no. 154, 2020. [2]     G. B. Moody and R. G. Mark, “The MIT-BIH Arrhyth- mia Database on CD-ROM and Software for Use with It,” Computers in Cardiology, vol. 17, pp. 185–188, 1990. [3]     T. R. Dawber, G. F. Meadors, and F. E. Moore, “Epidemi-ological Approaches to Heart Disease: The Framingham Study,” American Journal of Public Health, vol. 41, no. 3, pp. 279–286, 1951. [4]     World      Health      Organization,    “Cardiovascular diseases    (CVDs)  Fact Sheet,”WHO,  2022 [Online].Available: https://www.who. int/news-room/fact-sheets/detail/ cardiovascular-diseases-(cvds) [5]    M. B. Barla and A. K. Madam, “A Hybrid Deep Learn-ing Framework for Early Cardiovascular Disease Predic-tion: SDCN-CHDMO-ResVGG Approach,” Proc. 9th Int. Conf. Electronics, Communication and Aerospace Technology (ICECA), IEEE, 2025, pp. 1049–1053. [6]    Y. Wu, M. Huang, Y. Shi, Q. Li, Y. Zhang, and B. Song, “An Interpretable Hybrid Deep Learning Method for Heart Disease Prediction,” Proc. 5th Int. Conf. Digi-tal Society and Intelligent Systems (DSInS), IEEE, 2025, pp. 248–252. [7]     Ch. Vasavi and G. Sujatha, “Data Driven Heart Disease Prediction: A Comparative Study of ML Models,” Proc. 4th Int. Conf. Innovative Mechanisms for Industry Appli-cations (ICIMIA), IEEE, 2025, pp. 1711–1715. [8]     I. Sherly S., T. A. Mohanaprakash, T. Sindhu, N. Subra-manian, D. Bora, and K. Cinthuja, “Data Transformer – AI Driven Multimodal Fusion Network for Cardiovascu-lar Risk Prediction Using Multimodal Data,” Proc. 5th Int. Conf. Evolutionary Computing and Mobile Sustain-able Networks (ICECMSN), IEEE, 2025, pp. 203–212. [9]     L. Karthikeyan and E. Sathiyamoorthy, “Deep Learn-ing for Predicting Cardiovascular Events in Patients with Neurological Disorders: A Multimodal Analysis of Brain-Heart Interactions,” 2025 Innovations in Power and Advanced Computing Technologies (i-PACT), IEEE, 2025, pp. 1–5. [10]   C. Thirumarai Selvi, P. Seleena, R. S. Sankara Subraman-inan, AK. Shruthika, P. Gandhi Ayyavu, and S. Srinidhi, “Heart Disease Prediction using Feature-Optimized ML and Hybrid Deep Learning Models,” Proc. 3rd Int. Conf. Sustainable Computing and Data Communication Sys-tems (ICSCDS), IEEE, 2025, pp. 1025–1028.