Proposed Work- In this project we make use of the datasets provided by Physio-Net. The two datasets we are using are MIT-BIH Arrhythmia and PTB, they provide the ECG analysis for analysis. The aim is to train a deep neural network model that can learn the features that are present and predict the labels provided in the datasets with a high accuracy. The first step is to develop a model for the MIT-BIH dataset and to transfer this model to the PTB dataset using a machine learning model, called transfer learning. This method reuses a pre-trained model on a different model. It makes use of information it has learned from the previous assessment and uses that train the other dataset. It makes the process of developing another model faster instead of having to make a new model from scratch. This ideal model can make it simpler to train other models without having to start developing new models from scratch all over again.