Performance metrics with 1) all U-net layers trainable (Full), or 2) the first two layers (First), 3) the middle two layers (Middle), or 4) the last two layers (Last) (decoder) trainable, with the validation dataset.

The Std Dev Diff represents the standard deviation over validation sections of the differences of the metrics to the default Full Layer Trainable and the Avg Diff is the overall average differences compared to the default model. All models had a background weight of 1.0.