Thanks to all, I have the full text of the brain imaging paper and have had a look at it.
I have to say that, as for a lot of these studies, I am a bit puzzled by the results. I am also not very impressed by the introductory and concluding remarks. The authors talk of 'neuroinflammation'. To my mind a serious biomedical scientist or pathologist is not going to talk of 'neuroinflammation', which sounds to me like a lay-term concoction made to sound scientific. I would be interested in microglial activation or increased capillary permeability or cellular infiltration - all of which come under the general heading of inflammation - but I have never heard of some specific entity called 'neuroinflammation' and I am pretty sure it is a pseudo-idea.
The key point is that inflammation is a general term for a whole cluster of processes that has become rather old fashioned now that we know about the processes themselves. And if one of the processes occurs without the others that is not a 'sign of inflammation' so much as one of the processes without the others. So if this was full blown inflammation it could be picked up on ordinary MRI scans, or CT, and in fact it would be likely to produce specific clinical signs. If the thalamus is really inflamed you are usually unconscious or something pretty close to it. The authors talk as if neuroinflammation was something you get in multiple sclerosis and Parkinson's disease as well but I don't think there is any reason to think anything comparable is going on in the three conditions. It sounds to me like very muddled thinking.
Another point is that the hot spot on the scan in the ME case is very similar to the hot spot in the normal case but just a bit bigger. A normal brain should have no 'inflammation hot spot'. So it looks as if the authors are measuring something that you get in normal people to some degree and a bit more in ME. That would be interesting, but not inflammation. Moreover, the hot spot does not really look like an inflammatory response. It is the wrong shape. If you see a black object on the lawn you may think it is a crow, but if you then see it has four legs you change your mind. If this is an inflammation crow it looks to me like to have four legs.
And lastly, the authors give some very tight correlation patterns between scan readings in specific areas and symptoms but do not give a whole set of data - they just seem to have chosen the data that correlates well. That is the first mistake to make in statistical handling of data - to pick out the good bits and forget the rest. I would also like to see much more raw data in scatter plots so that one could see the real results rather than some predigested interpretation.
As for all these studies I am prepared to believe that I have underestimated, but there are certain issues here that make me reserve judgement. If this sort of detail does not look right then one always wonders about the main finding. The main finding of brighter hot spots in ME patients might indicate some change in brain function giving an important clue to what is going on, but calling it 'neuroinflammation' does not help and whatever it is normal people have about half the same thing going on.