Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Biomarker for chronic fatigue syndrome identified

med.stanford.edu

Post reply on HN
Print view View original

181–190 of 228 posts

Re: Biomarker for chronic fatigue syndrome identified

#181

Post by ufo »

sagebird wrote: post #153
kens wrote: post #137

Earlier quoted context omitted.

Thanks for the paper link. I read the paper and it's really a semiconductor sensor engineering paper, not a biology paper. The paper seems very preliminary; it shows that CFS and non-CFS sample have widely different impedances under osmotic stress, but the mechanism behind this is entirely hand-waving. If I'm reading the paper correctly, the samples were peripheral blood mononuclear cells (i.e. lots of different cell…

The use of machine learning needs to be explained. I’m half expecting it to boil down to “we evolved the neural nets until it perfectly separated the two groups” — overfitting “learning”.

I also had that impression at first but it seems that they are just using Support Vector Machines to draw the separating boundary between the two group of points.

Re: Biomarker for chronic fatigue syndrome identified

#183

Post by ufo »

leemailll wrote: post #179
ufo wrote: post #36

Earlier quoted context omitted.

I don't understand the details but here is the link to the paper https://www.pnas.org/content/early/2019/04/24/1901274116

Thanks for the link. I read the paper and it seems that they just measure the impedance (or resistance) of the cell membrane with an artificial increase of NaCl in the extracellular environment (ie plasma in the text). The final NaCl concentration looks like to be doubled than the normal range. This might be an intersting finding, but I can't figure out why the impdecance of membrane need to be measured with increase…

The salty plasma removes water from inside the cells by osmosis. The authors say that this technique is a common model for studying cellular response to stress.

Re: Biomarker for chronic fatigue syndrome identified

#184

Post by losteric »

p1necone wrote: post #171
everly wrote: post #134

Earlier quoted context omitted.

IMO he is clearly implying it came from a vaccine(s) but he is trying to do it without being seen as an anti-vaxxer.

Surely you'd have to have a lot of vaccines for it to compare to the amount you get from other sources (mostly seafood afaik?)

Yes. One vaccine contains less mercury than one (average) serving of fish, and the mercury is in a less toxic form.

Re: Biomarker for chronic fatigue syndrome identified

#186

Post by losteric »

whatshisface wrote: post #180
perl4ever wrote: post #178

Earlier quoted context omitted.

Well, there are lots of rare diseases, and mercury could be a factor in some of them. That doesn't mean they are imaginary. It's more of a definitional/philosophical issue, that some other factor is essential as well, so it's not right to attribute the problem only to mercury. The distinction I'm making is that I am pretty sure low level mercury is not an issue like lead from gasoline, although I can imagine evidence…

I'm also pretty sure that peanut butter is not an issue, but it does kill some people. Maybe there are people out there who are ultra-sensitive to mercury.

True, but people with deadly peanut allergies are still completely fine when peanuts aren't around... I can't imagine why vaccine mercury would explain life-long CFS.

Re: Biomarker for chronic fatigue syndrome identified

#187

Post by Smaug123 »

sagebird wrote: post #153
kens wrote: post #137

Earlier quoted context omitted.

Thanks for the paper link. I read the paper and it's really a semiconductor sensor engineering paper, not a biology paper. The paper seems very preliminary; it shows that CFS and non-CFS sample have widely different impedances under osmotic stress, but the mechanism behind this is entirely hand-waving. If I'm reading the paper correctly, the samples were peripheral blood mononuclear cells (i.e. lots of different cell…

The use of machine learning needs to be explained. I’m half expecting it to boil down to “we evolved the neural nets until it perfectly separated the two groups” — overfitting “learning”.

If you look at the graphs by eye, there is an extremely obvious divide. SVM just let them find it automatically, but if you gave a human those graphs, they could instantly point out where the line is.

Re: Biomarker for chronic fatigue syndrome identified

#188

Post by labster »

cheerlessbog wrote: post #176
tigershark wrote: post #152

Earlier quoted context omitted.

Do you eat 20 cans of tuna a day? When I was much younger I ate 1-2 cans of tuna pretty much every day for few years and I didn’t have any ill effect at all.

Did you just like the taste?

He's a tigershark, of course he likes the taste of tuna. Duh.

Re: Biomarker for chronic fatigue syndrome identified

#189

Post by feanaro »

tntn wrote: post #173
kens wrote: post #137

Earlier quoted context omitted.

Thanks for the paper link. I read the paper and it's really a semiconductor sensor engineering paper, not a biology paper. The paper seems very preliminary; it shows that CFS and non-CFS sample have widely different impedances under osmotic stress, but the mechanism behind this is entirely hand-waving. If I'm reading the paper correctly, the samples were peripheral blood mononuclear cells (i.e. lots of different cell…

This is less a criticism of the paper than it is a statement of misaligned expectations. > but the mechanism behind this is entirely hand-waving. The objective of this paper was not to specify, or even attempt to specify, the mechanism. > The paper's use of machine learning seems entirely pointless, basically drawing the obvious line between the two sets of points. Yeah, that is pretty much the entire objective of SV…

> Yeah, that is pretty much the entire objective of SVMs.

Which is hard to do by hand in many dimensions but almost trivial in 2. I haven't read the paper but maybe the OP is complaining about something like that?

Re: Biomarker for chronic fatigue syndrome identified

#190

Post by kens »

feanaro wrote: post #189
tntn wrote: post #173

Earlier quoted context omitted.

This is less a criticism of the paper than it is a statement of misaligned expectations. > but the mechanism behind this is entirely hand-waving. The objective of this paper was not to specify, or even attempt to specify, the mechanism. > The paper's use of machine learning seems entirely pointless, basically drawing the obvious line between the two sets of points. Yeah, that is pretty much the entire objective of SV…

> Yeah, that is pretty much the entire objective of SVMs. Which is hard to do by hand in many dimensions but almost trivial in 2. I haven't read the paper but maybe the OP is complaining about something like that?

Yes, the data is essentially 1-dimensional and looks like "CCC MMMMMMMM". (Seriously, look at Figure 2 graph H in the paper.) It seems entirely gratuitous to use machine learning to build a SVM classifier to separate the C's from the M's, when it's obvious that the C's are on the left and the M's on the right.
Post reply on HN