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

fMRI software bugs could upend years of research

theregister.co.uk

Post reply on HN
Print view View original

21–30 of 192 posts

Re: fMRI software bugs could upend years of research

#21

Post by Alex3917 »

JumpCrisscross wrote: post #12
Alex3917 wrote: post #7

It's already been known for several years that almost all MRI brain scan research is wrong, what exactly is new here?

> almost all MRI brain scan research is wrong Source for a layman?

Here are a couple Ben Goldacre articles on neuroscience in general:

https://www.theguardian.com/commentisfree/2011/sep/09/bad-sc...

For fMRI research specifically, I seem to remember that almost all of the analysis software is closed source, and that comparing the readings between software versions isn't actually meaningful even though a lot of papers do so.

Re: fMRI software bugs could upend years of research

#22

Post by acveilleux »

sxg wrote: post #10
jey wrote: post #5

Title should really read "fMRI" instead of "MRI". The referenced journal article is titled "Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates".

The same tools (AFNI, SPM, and FSL) are used in a lot of other MR research protocols as well - not just fMRI.

They are but 3dClustSim (the culprit) is essentially used only for fMRI.

Re: fMRI software bugs could upend years of research

#23

Post by acveilleux »

williamscales wrote: post #4

"Our results suggest that the principal cause of the invalid cluster inferences is spatial autocorrelation functions that do not follow the assumed Gaussian shape." In other words, researchers cut corners. You should never assume that something is a certain way without rigorously proving it. How did these papers make it past peer review?

fMRI papers tend to be written by neuro-psychologists or similar folks. Not engineers, statisticians, or physicists. They get reviewed by them too.

Re: fMRI software bugs could upend years of research

#24

Post by goalieca »

leemailll wrote: post #20
goalieca wrote: post #18

Earlier quoted context omitted.

I wasn't in fmri research but some fellow students in my lab were on it. I do know there was a study with dead salmon which were showing up as having "active" brain regions. I have to find it but am on my phone. It's easy to imagine the main difficulty with this is mapping signal to actual thoughts and regions. It's a really complex biology and physics to reduce.

what is your field? Neuroscience?

Electrical engineering (medical imaging - medical imaging technologies)

This is the best I could find for the paper. not what I originally meant to cite. http://onlinelibrary.wiley.com/doi/10.1002/scin.5591761320/a...

Re: fMRI software bugs could upend years of research

#25

Post by op00to »

Alex3917 wrote: post #7

It's already been known for several years that almost all MRI brain scan research is wrong, what exactly is new here?

It's not that the research is 100% wrong, but the way the research is reported in the news is generally wrong. No, researchers can't really tell if you're thinking about a sailboat from looking at an fMRI, despite whatever the newspaper says.

Source: worked in an fMRI research lab for 4 years.

Re: fMRI software bugs could upend years of research

#26

Post by greenyoda »

goalieca wrote: post #18
JumpCrisscross wrote: post #12

Earlier quoted context omitted.

> almost all MRI brain scan research is wrong Source for a layman?

I wasn't in fmri research but some fellow students in my lab were on it. I do know there was a study with dead salmon which were showing up as having "active" brain regions. I have to find it but am on my phone. It's easy to imagine the main difficulty with this is mapping signal to actual thoughts and regions. It's a really complex biology and physics to reduce.

Searching for "dead salmon fMRI" turned up the following:

http://www.prefrontal.org/files/posters/Bennett-Salmon-2009....

Re: fMRI software bugs could upend years of research

#27

Post by acveilleux »

Alex3917 wrote: post #7

It's already been known for several years that almost all MRI brain scan research is wrong, what exactly is new here?

There's a fair bit of junky data and quite a few analysis methods that cannot compare data accurately between magnets, position wrt isocentre, or even processing software revisions. But your claim is a lot bolder than that.

Re: fMRI software bugs could upend years of research

#28

Post by acveilleux »

Alex3917 wrote: post #21
JumpCrisscross wrote: post #12

Earlier quoted context omitted.

> almost all MRI brain scan research is wrong Source for a layman?

Here are a couple Ben Goldacre articles on neuroscience in general: https://www.theguardian.com/commentisfree/2011/sep/09/bad-sc... For fMRI research specifically, I seem to remember that almost all of the analysis software is closed source, and that comparing the readings between software versions isn't actually meaningful even though a lot of papers do so.

That's actually the same article on two sites. Itself just a layman's piece on an actual article in Nature Neuroscience.

Re: fMRI software bugs could upend years of research

#30

Post by acveilleux »

greenyoda wrote: post #26
goalieca wrote: post #18

Earlier quoted context omitted.

I wasn't in fmri research but some fellow students in my lab were on it. I do know there was a study with dead salmon which were showing up as having "active" brain regions. I have to find it but am on my phone. It's easy to imagine the main difficulty with this is mapping signal to actual thoughts and regions. It's a really complex biology and physics to reduce.

Searching for "dead salmon fMRI" turned up the following: http://www.prefrontal.org/files/posters/Bennett-Salmon-2009....

Back when I was involved in MRI research, multiple comparisons correction was the rock that many a paper in the field crashed upon. Neuro psych. suffers from weak stats, which is a far cry from saying either fMRI itself is meaningless or saying MRI as a whole is.
Post reply on HN