Myocardial imaging is a significant technique for diagnosing and checking coronary illness. Echocardiography is a normally involved imaging methodology for this reason. Profound learning calculations have shown promising outcomes in different clinical imaging undertakings. This review proposes a profound learning-based approach for myocardial capability imaging in echocardiography utilizing MLP and LR calculations. We utilized a dataset of echocardiographic pictures from patients with various sorts of coronary illnesses. The dataset was separated into preparing, approval, and testing sets. We pre-handled the pictures and applied information expansion methods to build the size of the preparation set. We prepared two distinct models: an MLP model, and an LR model. Our outcomes show that the MLP-based deep learning model beat the other model regarding precision, awareness, and explicitness. The MLP-based deep learning model accomplished a general exactness. The MLP model accomplished a general exactness and responsiveness of 0.91. Both calculations give the exactness higher than the current framework. The LR model accomplished a general exactness of 0.56. All in all, our review shows a profound learning-based approach involving a compelling device for myocardial capability imaging in echocardiography. This approach might work on the exactness of the conclusion and checking of coronary illness.