Yes it can but that is ok for restoring data, not for restoring a whole disk image.
The latter is not possible (with any OS)
Oh, it is quite possible, if you have the image's worth of unused storage space, and are using logical volume management. (Not just on Linux, but on many other OSes as well.)
You create a writable snapshot/fork of the current ("bad") filesystem, and restore the image over that snapshot, instead of the running system.
You can keep using your current system while the restore is in progress, too; the two are now separate. Many logical volume managers use copy-on-write, so stuff that does not diverge between the two does not need to be duplicated. (It might need a bit of help from the image tools; if they don't write the same data over the same block, COW is much easier to manage correctly.)
Before you reboot, or on your next reboot, you tell your volume manager which snapshot will be the active one. After you've booted to the new system, you can still examine the other one, and when you're done, delete it.
But before you ask, no, I don't know of any easy to use GUIs that can do this for you. I am comfy with the command interfaces, and have no idea what kind of GUI tools are available or not.