Parallel transmission (pTx) has proven capable of addressing two RF-related challenges at ultrahigh fields (≥7 Tesla): RF non-uniformity and power deposition in tissues. However, the conventional pTx workflow is tedious and requires special expertise. Here we propose a novel deep-learning framework, dubbed deepPTx, which aims to train a deep neural network to directly predict pTx-style images from images obtained with single transmission (sTx). The feasibility of deepPTx is demonstrated using 7 Tesla high-resolution, whole-brain diffusion MRI. Our preliminary results show that deepPTx can substantially enhance the image quality and improve the downstream diffusion analysis.