Kahneman's work is really interesting, but don't think it's that relevant here. It's not a situation
of "if it saves only one life...". The Imperial College's study presents very clear outcome predictions for the different options available at this point. They might be wrong, but this isn't an area where semantics make that much of a difference.