alt Hacker NewsIt was clearly wrong for them to be immediately dismissive of the claim (particularly since there are a number of existing studies that support it).
> I guess a blind study with a control group wouldn't be too hard to conduct.
However, you would be surprised how difficult this is to do. As soon as you are doing any sort of research involving human beings, even if that research is plainly innocuous for the subjects and technically easy to conduct, everything gets very complicated due to regulations meant to prevent misconduct and abuse.
I agree, but even with all the authorizations, it looks like a hard problem. You need like 20 persons with cancer. Before starting chemo? A matching group of 20 healthy persons. What about symptoms of cancer, like big moles in skin cancer or low weight in colon cancer?
Since the person with/without cancer know the situation, they may give involuntary hints. Just sedate them so they are sleeping while testing?
What if the person can detect some cancers and no others?