The rising prevalence of cardiovascular disorders has led to a substantial improvement in their diagnosis. Despite the fact that several techniques have been created for illness classification and privacy protection for safe data transfer, most of these techniques still have problems with precise decision making. In the field of clinical data investigation, the expectation of cardiovascular activity is a fundamental test. Artificial intelligence (AI) has matured into a useful tool for sifting through the mountain of data produced by medical professionals and health organisations to make informed decisions and forecasts. Machine learning techniques are helping doctors better predict heart attacks, and this has led many people to alter their behaviour in the future to avoid a similar fate. This study looks at two data sets, one from UCI and one from Kaggle, to make predictions about cardiovascular illness. The results obtained by applying the suggested model to these two datasets are competitive with current approaches. The heart disease prediction model presented here makes use of a Dual path network with recurrent convolutional cascades. Both the individual channel and the inter-channel connection are taken into account when employing LSTM. Then, layers that are completely interconnected occur. For this reason, the proposed system outperforms preexisting smart heart disease prediction systems by a wide margin, with an accuracy of 91.7% and 93.7%, respectively, when tested on the UCI and Kaggle datasets, respectively.