Lecture 3 Flashcards

Question

why do we use an integrated neuroscience approach?

Answer 1

think about how brain may change in response to diff functions of life

Answer 2

tissue classification, white-matter density, deformation field, segmentation, cortical thickness, cerebral sulci

Answer 3

choice of acquisition protocol (FLASH, MPRAGE, MDEFT) -> differences in contrast

Answer 4

since u manipulate water, lot of free parameters u can use same person scanned with three diff parameters same algorithm set to work on mri data can give different answers based on which parameter used to acquire it (keep in mind)

Answer 5

need common coordinate frame -> how homologous anatomy is over time — need way to create this reference system

Answer 6

initially derived by aligning 300 brains and saying that this is the normative anatomy of the brain can then look at mean morphology

Answer 7

1. principle is the same — you get volume from scanner and want to align it to a space 2. normalize it across scans 3. do some linearization/transformation 4. use algorithm to classify the grey and white voxels 5. smooth out and by voxel/voxel level, run stats (color map — high = lot of difference)

Answer 8

surface representation is just points in space - can go and measure differences btw white and black matter surface - can then put this in a color map algorithm creates mesh - see points that define the surface (edges of triangles) - ask algorithm to fold the sphere to match the voxel representation of the white matter - constantly creating more points that allow it to fold more - big brain morphologies are the ones done earlier on

Answer 9

one fold increase in terms of brain volume in rats housed in the complex house

Answer 10

made people learn to juggle scan1-2 increase in grey matter (region V5 —> regions is like motor and visual) sees rapid growth (3% change) in this part of the brain over time then asks people to stop juggling and three months later → this part of the brain is still enlarged compared to baseline

Answer 11

differences in brains of musicians (uni level) vs control see increased cortical thickness in PFC and regions involved in working memory and involved in sound processing representation is far more compact (not diffuse) suggests that years of training have specific imprint on the brain that doesnt exist in non-musicians

Answer 12

brain changes a lot in early life things that change are grey matter, myelin, white matter concentration brain is continuously refining across lifespan white matter conc peaks in end middle age

Answer 13

brain matures later in adhd kids in PFC there's almost a 3y delay of reaching cortical thickness as normal developing kids

Answer 14

1. parietal-frontal progression of GM loss during adolescence (from age 12-16) 2. GM loss declines and is circumscribed to prefrontal and temporal cortices by age 24

Answer 15

healthy COS siblings show cortical GM deficits in early ages that NORMALIZE by age 18

Answer 16

heterogenous; many factors contribute

Answer 17

prevalence: 1 in 68 The ratio of males to females diagnosed is about 4:1 (more pronounced in higher functioning (6-9:1), but closer in lower functioning (2:1))

Answer 18

female protective effect hypothesis

Answer 19

States that females may be protected against neurodevelopmental disorders May take more autism risk factors (genetic or environmental) to manifest the autistic phenotype Females have higher threshold for the amount of genetic liability or risk factors needed to cause autism (dotted line), so affected less often than males Supported by the fact that, presumably due to this higher burden or number of risk factors, when females are affected, tend to be more severe

Answer 20

Other studies have examined the neuroanatomy of autism into development and adulthood and found significant effects, in particular cortical thickness But inconsistent: many studies have found greater thickness in individuals with ASD, but others have found thinner cortex or no differences. age seems to be more important: found greater CT but effects normalize with age

Answer 21

may have moderating effect: CT positively correlated with IQ in controls, but not in ASD (no or negative relations)

Answer 22

Participant motion is: Associated with reduced average cortical thickness Higher in clinical populations More strongly correlated with CT than volumetric estimates

Answer 23

Assign motion score (4 point scale) Two independent raters of motion → QC manual, trained raters, checked reliability after first 100 scans → Consensus score Objective exclusion cut-off looking for ringing and blurring

Answer 24

Random effects meta-analysis used to account for inter-site differences and heterogeneity of sample Multiple linear regression across 80,000 vertices conducted within each site Cohen’s d effect size calculated per site at each vertex and pooled in meta-analysis False Discovery Rate (FDR) for multiple comparisons correction

Answer 25

allows to treat sites as separate studies and analyse that way to avoid confounds of scanner and variability of demographics between sites

Answer 26

First, we looked at the main effect of DX in all subjects, controlling for age and sex, and found widespread increases in CT across in the brain in ASD These are maps of q-values across the brain, regions where significant diffs between ASD and controls - all increases in ASD All coming slides are 5 or 10% FDR, and min 1%, all increases in ASD

Answer 27

We then broke this down to look at males and females separately, and here is what we found in the males -- shown at 10% FDR Again, increases in CT, primarily in inferior frontal and superior temporal regions. Finally as cubic, effects even stronger, still strongest at 8, decreasing at 12 and 16, still small areas of significance visible at 20 (drop after 40)