Deep Learning-based magnetic resonance imaging (DL-MRI) could accelerate 3D MRI scanning time. In this study, we investigate the feasibility of DL-MRI in 2D shoulder MRI. Totally 20 consecutive patients were enrolled for both conventional MRI and DL-MRI. Both qualitative and quantitative analyses were conducted to compare the image quality and lesion diagnosis on conventional MRI and DL-MRI. And our results revealed that DL-MRI was valuable for improving the overall workflow of shoulder MRI with scanning time saved and image quality improved.