Why we need a CTO Image?

If a CTO node was smartfailed, it was removed from the cluster.

- Why we need this special CTO image to re-image the node?

- Why we can not use the standard image or better, what if the problem if we use the standard image and join the node to the cluster?

If we receive a new node from the factory, it has the standard image on it. What is the difference here?