Fig 5. Investigation on the generalization capability and over-fitting possibility of the CNN model.

Input and predicted images by 64- and 16-channel sparse array, and the ground truth images by the original 128-channel dense array of (a) a tooth from a new porcine jaw that was separate from the training dataset, and (b) radial artery. (c) Input and predicted images by 64- and 16-channel sparse array of noise signals.