Skip to main content
. Author manuscript; available in PMC: 2023 Jul 25.
Published in final edited form as: J Surg Res. 2022 Nov 24;283:500–506. doi: 10.1016/j.jss.2022.10.069

Figure 2:

Figure 2:

Figure shows the workflow architecture of the deep learning network. Frames from the video clips for each of the basic maneuvers are extracted and fed into a convolutional neural network and an LSTM. The architecture is then trained to predict the maneuver based on an input video clip.