Focused Assessment with Sonography in Trauma (FAST) Ultrasound is a medical procedure to assess for free fluid following physical trauma. These ultrasound images can often be difficult to interpret and requires operators to be appropriately trained. Traditionally, skill is assessed by direct observation from experts, which is expensive and error prone. This project aims to use deep learning to provide automated skills assessment for FAST exams. A set of FAST ultrasound videos of each vital region (perihepatic, perisplenic, pericardial, pelvic) performed by 7 novice users, 7 intermediate users, and 3 expert users was used. A modified I3D network, a modern neural network with a focus on action-based items, was retrained to first identify the region being scanned, and then identify the skill level of the user from the ultrasound videos. The model’s performance was evaluated using k-fold hold-out. Results found a testing accuracy of 28.57% for region identification and an accuracy of 71.43% for skills assessment using the I3D network. These results were compared with previous results on the same dataset of 39.09% and 70.00% for the respective tasks. These results are an improvement over the previous results for skill level evaluation, implying potential use of an I3D network for evaluating skill level in the future with the proper finetuning. More work is required to accurately identify region.