Advanced Heart Disease Prediction Through Spatial and Temporal Feature Learning with SCN-Deep BiLSTM

• DOI: 10.1007/s44196-025-00734-6
• Publication Date: 2025-02-10T13:40:16Z
• Authors: Vivek Pandey; Umesh Kumar Lilhore; Ranjan Walia
• Keywords: Feature (linguistics)

Heart disease prediction using machine learning methods faces various challenges, such as low data quality, missing irrelevant values, and underfit overfit problems, which increase the time complexity degrade model's performance. Moreover, hybrid models for heart showed poor accuracy due to irrelevancy in dataset. Therefore, a search optimizer with deep convolutional neural network coupled Deep Bidirectional long short-term memory classifier (SCN-Deep BiLSTM) is proposed handle abovementioned issue. The importance of SCN-Deep BiLSTM relies upon establishing spatial information temporal features from ECG signals that support while minimizing computational associated raw signals.The model achieves accuracy, F-score, precision, recall, critical success index 0.97, 0.98, 0.99, respectively 80% training, whereas attained 0.96, 0.94, 0.96 recall index, when K-Fold 10. performance outcome emphasizes efficacy accurate classification disease.