Hi Stijn,
I will carefully read your paper. In the meantime, a quick answer to your
questions.
My data about brain atrophy are corrected for age, education, handedness
and ICV. All subjects are female. GM is decreased because of prolonged and
severe undernutrition in AN and is usually reversible. So I think it is a
bit different situation from aging.
I am studying connectivity using resting state fMRI and I have used ICA
and dual regression.
My question is if there are other situations where an increase in
connectivity (functional not structural) is present in atrophic areas of
the brain.
AN and control groups differed significantly on this and the connectivity
of the AN group significantly correlates (on a voxel basis) with the total
amount of weight loss (a negative correlation!). A similar correlation is
present in healthy subject, but it is unsignificant at cluster-based
correction.
Thank you for your interest and let me know what you think about it
Angela
> Hi Angela,
>
> Interesting to have a topic on interpretation of results. It's always good
> to share some knowledge and discuss. I did some research on healthy brain
> aging.
>
> Your analysis is on gray matter in various area's as you state. First you
> notice a significant loss of gray matter in the patient group, which is
> interesting. Are you sure about matching your groups on age, gender,
> handedness, education, IQ and such? Variables may influence your results.
> Another question is on normalizing your results. Different subjects have
> different intra-cranial volume (ICV). To make sure you compare correctly,
> you should normalize the specific areas to the ICV.
> It is known that gray matter volume decreases in healthy aging (and
> cerebrospinal fluid increases). Different structures in the brain age
> differently. There is a theory based on the frontal-occipital way of aging
> where the frontal part ages quicker than the occipital part. Consider
> checking my paper on healthy aging:
> http://www.ncbi.nlm.nih.gov/pubmed/20483378
>
> How do you define the connectivity in the GM areas you investigated?
> Usually connectivity (like FA, MD, ADC) is investigated in white matter.
> In this respect I can't say what's right. Your finding might be a
> compensation mechanism of the brain, getting less matter but 'better'
> connections. If you assume that the increase of connectivity is presented
> because of the atrophy, you might check for a relation with illness
> duration/age of onset/severity. A hypothesis like "A longer the illness
> duration makes an increase of connectivity" can be used. And is the
> increase in connectivity significant between your two groups?
>
> Let me know what you think!
>
> Kind regards,
>
> Stijn Michielse
> Research Assistant
> Dept. Psychiatry and Neuropsychology
> Maastricht University
> E-mail: [log in to unmask]
>
> ------------------------------------------------------------------------------------------
> Hi FSL masters,
> I (again) need some advice on analyses/interpretation of my data.
> I am analysing a sample of patients with atrophy of GM in various brain
> areas due to malnutrition and weight loss (it is a sample of anorexia
> nervosa subjects). I have used ICA and dual regression to identify
> networks and analyse differences between patients and controls.
>
> I have found an increase in connectivity in areas that (on a voxel based
> morphometry) have a significant loss of GM.
> Is this a contradictory finding? Is it possible to observe an increase of
> connectivity as a 'compensation' of brain atrophy? Or may I consider this
> finding as an increase of connectivity due to the psychiatric illness
> observable despite atrophy?
> It is not easy to disentangle the effects of weight loss, brain atrophy
> and what is due to (or at the origin of) the psychiatric disorder
>
> I know that interpretation is up to me (and my clinical data), but I am
> wondering if something similar has been observed in other patients
> populations with brain atrophy.
>
> thank you for any help!
>
> Angela
>
> Angela Favaro, MD, PhD, MSc
> Psychiatric Clinic
> Department of Neurosciences
> via Giustiniani 3
> 35128 Padova
> Italy
>
>
|