Predicting a person’s chronological age based on their neuroimaging data is an important biomarker for early detection of Alzheimer’s disease and an essential step toward better understanding the normal development of the brain. In this paper we present an efficient and fast deep learning approach for predicting an individual’s brain age using T1-weighted 3D brain MRI scans. Whole brain MRI is preprocessed using MATLAB for extracting the main features which is directly related to aging and then applied augmentation techniques to increase the diversity. The proposed model was trained using 1126 T1-weighted MRIs from IXI public dataset. This approach achieved better performance along with less computations as the proposed model was lightweight than many 3D based convolutional neural networks for brain age prediction and traditional machine learning methods.