Delete and start over?

[Kupuna]

I mistakenly scanned in a backup copy of my major MM database. I now have nearly 800 duplicate files. I've used the find duplicates command but problem is (for example) that some of my current db entries have been converted from m3p to m3a. They properly show up as duplicates, but I have to wade through choosing the "best" track in each case.
My plan is to delete the entire current database and then re-scan the most current stored db containing the tracks. I've read that I will lose certain data details, but compared to handling 800 duplicates one by one, this seems the cleanest way to go. Does anyone agree or have alternate tactics?
SECOND QUESTION:
Once I get the db back in shape, is there any command or add-on that would flag or prevent adding a duplicate track? Not after I've done it, but as I'm trying to load the track? Probably not, but it would be a great idea.

[Stigg]

Enable the "Added" column, then sort by it by clicking on it so all the recently added tracks come together.
Then simply highlight and remove them.

Disable the "Added" column.

[Lowlander]

You can also use the Location sub-node to delete files from the Backup location.

There is no option to prevent duplicates from being added on scanning. You can manage them afterwards: http://www.mediamonkey.com/support/inde ... ate-tracks