This is the entirety of Nicomachean Ethics VI.3:

Let us begin, then, from the beginning, and discuss these states once more. Let it be assumed that the states by virtue of which the soul possesses truth by way of affirmation or denial are five in number, i.e. art, scientific knowledge, practical wisdom, philosophic wisdom, intuitive reason; we do not include judgement and opinion because in these we may be mistaken.

Now what scientific knowledge is, if we are to speak exactly and not follow mere similarities, is plain from what follows. We all suppose that what we know is not even capable of being otherwise; of things capable of being otherwise we do not know, when they have passed outside our observation, whether they exist or not. Therefore the object of scientific knowledge is of necessity. Therefore it is eternal; for things that are of necessity in the unqualified sense are all eternal; and things that are eternal are ungenerated and imperishable. Again, every science is thought to be capable of being taught, and its object of being learned. And all teaching starts from what is already known, as we maintain in the Analytics also; for it proceeds sometimes through induction and sometimes by syllogism. Now induction is the starting-point which knowledge even of the universal presupposes, while syllogism proceeds from universals. There are therefore starting-points from which syllogism proceeds, which are not reached by syllogism; it is therefore by induction that they are acquired. Scientific knowledge is, then, a state of capacity to demonstrate, and has the other limiting characteristics which we specify in the Analytics, for it is when a man believes in a certain way and the starting-points are known to him that he has scientific knowledge, since if they are not better known to him than the conclusion, he will have his knowledge only incidentally.

Let this, then, be taken as our account of scientific knowledge.

That is to say, science is systematic, demonstrative knowledge of what is necessary. (It excludes first principles; it only includes what is derived from first principles.)

Not only is that not modern science (and not what philosophers discussing the problem of demarcation are discussing) but it isn't clear where one would fit modern science among Aristotle's states by virtue of which the soul attains truth.

All that is to say that I think philosophy of science is pretty wide open for Aristotelians. Feser tends to lean toward epistemic structural realism. But it isn't obvious that any particular view is forced on one by Aristotelianism. I'd think an Aristotelian "could" be a selective realist, or a constructive empiricist, or a Kuhnian. (Each of these views, of course, have their own merits, and may have to be qualified to varying extents to fit with Aristotelianism, e.g., selective realism with the doctrine of virtual presence).

One finds among some Thomists (e.g., William Wallace and some more popular writers) the idea that the empirical sciences provide data on which philosophy can judge, so that by systematizing and interpreting the results of empirical sciences one might arrive at a completed Aristotelian science, i.e., demonstrative body of knowledge.