If you have a loved one in the hospital with a terminal illness and this person
What can be considered ethical will, of course, depend on what conception of ethics (and also, perhaps, what metaphysical assumptions) one brings to bear on the question. So, for example, if one believes in the absolute sanctity of (human) life, then the fact that the loved one continues to be (biologically) human and alive will provide the answer. On the other hand, if you think that what makes something human (in the ethical sense) is the ability to make decisions (I think it can be argued that Aristotle, for example, believed this), then it would seem that it is ethically open to end the biological life of the loved one, since their human life had already ended, in your scenario. However, it would also follow from the Aristotelian view that there is nothing wrong with terminating the life of a fetus (or even infant!) until the time it is able to engage in decision-making.
In other words, a proper answer to your question must flow from having, first, an adequate conception of what sort of being we are dealing with here. The fact that something is biologically human and alive does not seem to be enough to make a confident ethical judgment--we wouldn't, for example, afford a "right to life" for living tissue samples, for example. Once we get a grip on that question, then we also need to make some basic ddecisions about how things of that kind are to be recognized within our ethical theories.
Speaking just for myself, I am inclined to be very suspicious of "all or nothing" answers to these sorts of questions, as if there is some clear conceptual breaking point between what deserves to be afforded a right to life and what does not. But certainly ethical theories that recognize a central role for "rights" not only exist, they continue to flourish within the philosophical community.
In the ethical approach I favor (virtue theory), questions about when and what circumstances merit which sorts of actions are understood, instead, in terms of desirable characteristics of agents. In this way of looking at things, your sort of question would turn into something like this: What would an excellent human being decide to do about a loved one in such a circumstance? I expect that the questioner might find this version of the question not much help, because it is difficult to try to get a grip on all that an excellent human being might bring to bear on such a decision. So, virtue theorists, I think, have to concede that there may be no such thing as fixed and fully general answers to such questions--nothing that might be regarded as an "algorithm" or decision principle for making ethical judgments of the sort you seek here. That may be simply a failing of virtues theories, and some critics have held it to be such. But some of us, anyway, think that it is actually a virtue of virtue theory--it recognizes the necessity of the exercise of human judgment, because the morally relevant circumstances of any given situation may be such as to merit different answers in different cases.
In other words, your question seems to want to force a "yes" or "no" answer, and I am suggesting that either answer is likely to be artificial and forced. Does the loved one have medical insurance that will indemnify his or her heirs against the likely heavy medical costs of continuing his or her life? Are there other loved one's whose opinions and feelings about this very question matter, and if so, how do they feel about it? Would the limited resources that are going to be used in maintaining the life of this person be such as to be better assigned to something (or someone) else? Did this person leave a living will, or some other document that proposed an answer to this very question, if such a situation arose? And so on... I suspect that if we tell the rest of the story in such a way as to capture all of the morally relevant details, we might find we are prepared to offer different answers to different cases, precisely because of the effects (potentially incommensurable, but still significant to the agent practicing judgment) of these morally relevant details.
In brief, then, I think you should be wary of the very idea that questions like this one can be given clear, definitive, and ethically defensible answers! Not so fast!