Discussion in 'Other Health News and Research' started by Kyla, Jun 15, 2016.
I'd like to know if this test, or similar, has been used to test ME patients for auto-antibodies...
Rough and grumpy today, so sounding off accordingly:
With 9,486 different proteins, that's very, very likely to be a false positive. No sign of appropiate statistical corrections in the paper's abstract Detection of Alzheimer's disease at mild cognitive impairment and disease progression using autoantibodies as blood-based biomarkers
(though see this quote on the approach they used, from wiki "Random decision forests correct for decision trees' habit of overfitting to their training set.:587–588" edit, correction: but still a small sample and 9,000+ variables to pick from...).
No, they don't.
IF it gets past that stage, it might be interesting.
@Simon does this work have a hypothesis?
You can also try a Google Site Search
Separate names with a comma.