Skip to main content
. 2021 Oct 29;45(2):zsab260. doi: 10.1093/sleep/zsab260

Figure 1.

Figure 1.

Visual data flow of our experimental pipeline.

(A) A visual description of how we organized data collection, annotation, feature generation, and classifier training. (B) A visual description of frame-level information we use for visual features. We used a trained neural network to produce a segmentation mask of pixels pertaining to the mouse for use in downstream classification.