Heart Blockage Prediction Using Differential Equation with R-LOOKAHEAD- LSTM and DWARF Mangoose Optimization Based on Squeeze Net Method

Abstract

Huge diversity has evolved in the medical industry as a result of improved computer power and technology, particularly in the detection of cardiac occlusive disorders in humans. It has a devastating effect on human life and is now one of the worst cardiac illnesses in humans. To increase the model's capacity for prediction, this study suggest an optimization approach for dwarf mongoose and using differential Equation with R-Lookahead-LSTM-based cardiovascular disease prediction model. Using the R-Lookahead method, the long-term memory network model was further optimized. It is possible to enhance LSTM network models to demonstrate the model's propensity for speed and stability. This model can then be used to forecast cardiovascular disease. Additionally, we employ Squeeze Net adjusted using the R-Lookahead-LSTM-DMOA dwarf mongoose optimization method, which modifies several components of dwarf mongoose feeding to anticipate cardiovascular disease. This research includes a collection of ECG signal data from individuals with cardiac conditions. This unique dataset consists of 1937 individual patient records, all of which were gathered using an ECG Device. The data acquired comprises 12 leads-based ECG signal data. Additionally, according to the experimental findings, the R-Lookahead-LSTM-DMOA Squeeze Net method's highest accuracy is 0.867287, its precision is 0.846625, and its recall is 0.896720.