short answer, no
A big red flag in database design are comma seperated values.
and scanning through each row's numbers and comapring them to the same tables numbers can be quite a intensive task since you are dealing iwth strings.
your best solution is to break up the comma seperated values into it's own rows.
I use to think csv were a good solution too, but you actually dont save any space or speed and you lose quite a bit of relational functionality.
so I would put fourth your best effort into the table redesign, but if you MUST stick to csv then I suppose to could do a query, sub query using LIKE for 1 specific value, if you are looking to associate the two then you are looking a million of combiations after a few dozen to a few hundred rows