. High-throughput, fully automated volumetry for prediction of MMSE and CDR decline in mild cognitive impairment. Alzheimer Dis Assoc Disord. 2009 Apr-Jun;23(2):139-45. PubMed.


Please login to recommend the paper.


  1. I must first declare an interest as I am a member of the ADNI MRI core.

    This is an interesting if largely confirmatory paper. The authors show that medial temporal lobe atrophy in mild cognitive impairment (MCI) is associated with subsequent cognitive decline. The paper brings to bear the power of ADNI in terms of 1) large numbers of subjects, 2) freely downloadable serial imaging, and 3) an easily accessible comprehensive database of clinical, cognitive, and biomarker information. This paper constitutes a relatively large study although it uses a fraction of the ADNI data: namely, just the baseline MRI of 269 MCI subjects, baseline cognition, and change in MMSE and CDR over the following six months. The authors showed that smaller hippocampal and amygdala volumes and larger ventricular size (temporal horn) were associated with clinical decline. The first reports of the association between medial temporal lobe atrophy on MRI and Alzheimer's disease (AD) date back over two decades now. (As an irrelevant aside, one of those very first reports, Seab et al., 1988, was co-authored by one of the current ADNI investigators, Bill Jagust.) Over the last 20 years a number of papers have shown that AD and progression to AD is presaged by medial temporal atrophy on MRI, whether this is assessed by simple visual rating or by painstaking manual hippocampal tracings. What is interesting in this paper is the fact that the measures were fully automated (using commercial software). Automated software offers potentially very significant savings of time and effort in drawing round structures like the hippocampus on MR scans—and makes conceivable the idea of measuring several different structures on large numbers of scans. Ideally such software would be freely available—like the ADNI data. Amidst the bewildering number of studies using the ADNI data, we will learn what are the most effective measures of predicting clinical conversion from MCI and of tracking progression. These data facilitate head-to-head comparisons of the various markers. Hopefully they will also stimulate the development of new methods and new algorithms that will be available to improve clinical and research assessments in aging, MCI, and AD—and the prediction of change along that continuum.


    . Quantitative NMR measurements of hippocampal atrophy in Alzheimer's disease. Magn Reson Med. 1988 Oct;8(2):200-8. PubMed.

Make a Comment

To make a comment you must login or register.