Ensemble Learning is a technique that combines several individual models to create a stronger and more powerful model than any of its individual components. This method also helps reduce the risk of overfitting due to the diversity of the models being combined. Ensemble learning has been successfully applied in many fields and often outperforms single models. There are various ensemble techniques, such as averaging, bagging, random forest, stacking, and boosting. These techniques differ in how they train and combine the individual models. This chapter discusses application of deep learning ensemble architectures for medical applications.