On August 8th when I posted my review of the February 10, 2010 Journal of Neuroscience article that used Alzheimer’s Disease Neuroimaging Initiative (ADNI) repository data to look for biological markers that would predict the onset of Alzheimer’s disease, I had no idea that the very next day the New York Times would post a sensational story about a new research article based on ADNI data.

The last sentence in the results section of the article’s abstract states that in a set of 57 patients “with mild cognitive impairment followed up for 5 years, the model showed a sensitivity of 100% in patients progressing to Alzheimer’s disease.” This was the “100 percent accurate” cited by the New York Times and numerous posts over the Internet. “The model” that they refer to is their unique way of using a certain mixture of amyloid beta and tau proteins in the patient’s cerebrospinal fluid as the biomarker rather than each protein independently.

It’s exciting to see papers pouring forth based on large online data repositories where scientists are able to analyze much more data than any one team could collect on their own. In the new article the authors seem to have applied an innovative approach to analyzing the data set.

The new article was just published in the August 2010 issue of Archives of Neurology and I have not managed to get access to the full text yet. When I do I will post a review.


Other related blog posts:

Brain Research Using Online Data Repositories: Predicting Alzheimer’s Disease