One of the challenging medical areas that makes it difficult to predict an event before it occurs is a cardiac attack or syndrome. Insufficient oxygen intake will result in improper electric heart rhythm balance, changes in irregular heartbeats over time, and imbalance in all bodily functions throughout the day, which will ultimately cause an attack of heart valves and blood flow. Since the disease has a very slow response time, early-stage prediction is necessary. The main goal of the study is to create an expert system that uses multi-variate feature predictors and Deep Learning (DL) classification algorithms to handle independent and multi-class variables to predict the incidence of a cardiac attack for a patient at the current stage or in the future. The research aims to create a framework model using cutting-edge algorithmic techniques that will motivate medical professionals to find early warning signs of cardiac attack so they can find ways to save patients. Around fifty reviews are essential for designing the architecture of the expert system design for the prediction of a cardiac attack according to a thorough literature re- view on different algorithms and their effectiveness in predicting cardiac attacks. This knowledge review proposes may lead to addressing early cardiac attacks that accept the selected features from different feature predictors and forms a fuzzification of selected features along with rule sets and knowledge-base from experts using the DL and data mining algorithms.