This study would be much more believable if they had used pre- and post-infection scans from the same people instead of artificially constructing the comparison

This study would be much more believable if they had used pre- and post-infection scans from the same people instead of artificially constructing the comparison groups. The study needs to be carefully analyzed by data scientists; it’s just too easy to cherry-pick the groups and claim to have found a significant difference.