This Brain age prediction is a novel biomarker approach wherein the biological age of an individual's brain is predicted from neuroimaging data. Currently, most of the state-of-the-art predictions are made from T1-weighted MRI scans, and accuracy has been improved considerably since recent developments in deep learning with convolutional neural networks and their 3D architectures. This review covers a deep learning-based study on brain age prediction regarding pre-processing of data, model architecture design, evaluation metrics, and practical challenges such as dataset bias, interpretability, and cross-dataset generalization. The results have underlined the clinical potential of brain-age delta as an early marker of cognitive decline and neurological disorders, while pointing to robust training strategies and reliable methods for proper real-world validation.