Differences between revisions 5 and 6
Deletions are marked like this. Additions are marked like this.
Line 1: Line 1:
Line 3: Line 4:
= Two Groups (One Factor/Two Levels), One Covariates = ~+'''Two Groups (One Factor/Two Levels), One Covariate'''+~
Line 5: Line 6:
This models the input as two separate lines (DODS), one for each
group. The two groups can be thought of as two levels of a single
discrete factor. The covariate can be thought of as a continuous
This models the input as a single line (ie, an intercept and a slope).
Line 10: Line 9:
=== FSGD File (g2v1.fsgd) === <<TableOfContents>>

= FSGD File (g1v1.fsgd) =
Line 14: Line 15:
Class Group1
Class Group2
Class Main
Line 17: Line 17:
Input subject1 Group1 30
Input subject2 Group2 40
Input subject1 Main 30
Input subject2 Main 40
Line 21: Line 21:
Nclasses = 2 <<BR>> Nclasses = 1 <<BR>>
Line 24: Line 24:
=== Regressors (DODS) === = Regressors =
Line 26: Line 26:
Nregressors = Nclasses*(Nvariables+1) = 2*(1+1) = 4 <<BR>>
Regressor1: ones for subjects in Group 1, 0 otherwise. Codes intercept/mean for Group 1 <<BR>>
Regressor2: ones for subjects in Group 2, 0 otherwise. Codes intercept/mean for Group 2 <<BR>>
Regressor3: age for subjects in Group 1, 0 otherwise
. Codes age slope for Group 1 <<BR>>
Regressor4: age for subjects in Group 2, 0 otherwise. Codes age slope for Group 2 <<BR>>
Nregressors = Nclasses*(Nvariables+1) = 1*(1+1) = 2 <<BR>>
Regressor1: All ones. Codes intercept/mean for Main <<BR>>
Regressor2: age for each subject. Codes age slope for Group 1 <<BR>>
Line 32: Line 30:
=== Contrasts === = Contrasts =
Line 41: Line 39:
===== Contrast 1 =====
Null Hypothesis: is there a difference between the group intercepts? Is there a difference between groups regressing out the effect of age?
== Contrast 1 (intercept.mtx) ==
Null Hypothesis: is the intercept equal to 0?
Line 44: Line 42:
Contrast File: g1-g2.intercept.mtx
Line 46: Line 43:
1 -1 0 0 1 0
Line 49: Line 46:
This is a t-test with Group1>Group2 being positive (red/yellow). This is a t-test with the intercept>0 being positive (red/yellow).
Line 51: Line 48:
===== Contrast 2 ===== == Contrast 2 (slope.mtx) ==
Line 53: Line 50:
Null Hypothesis: is there a difference between the group slopes? Note:
this is an interaction between group and age. Note: not possible to
test with DOSS.
Null Hypothesis: is the slope equal to 0?
Line 57: Line 52:
Contrast File: g1-g2.slope.mtx
Line 59: Line 53:
0 0 1 -1 0 1
Line 62: Line 56:
This is a t-test with Group1>Group2 being positive (red/yellow). This is a t-test with the slope>0 being positive (red/yellow).
Line 64: Line 58:
===== Contrast 3 =====
Null Hypothesis: does Group1 differ from Group2 in intercept or slope?

Contrast File: g1-vs-g2.mtx
{{{
1 -1 0 0
0 0 1 -1
}}}
Note: this is an F-test (and hence unsigned). Reversing the signs will
have no effect.

===== Contrast 4 =====
Null Hypothesis: does mean of group intercepts differ from 0?

Contrast File: g1g2.intercept.mtx
{{{
0.5 0.5 0 0
}}}

This is a t-test with (Group1+Group2)/2 > 0 being positive
(red/yellow). If the mean is < 0, then it will be displayed in
blue/cyan.

===== Contrast 5 =====
Null Hypothesis: does mean of group slopes differ from 0? Is there an average affect of age regressing out the effect of group?

Contrast File: g1g2.slope.mtx
{{{
0 0 0.5 0.5
}}}

This is a t-test with (Group1+Group2)/2 > 0 being positive
(red/yellow). If the mean is < 0, then it will be displayed in
blue/cyan.

=== mri_glmfit command ===
= mri_glmfit command =
Line 106: Line 65:
  --glmdir g2v1 \   --glmdir g1v1 \
Line 108: Line 67:
  --fsgd g2v1.fsgd \
  --C g1-g2.slope.mtx \
  --C g1-vs-g2.mtx \
  --C g1g2.
intercept.mtx \
  --C g1g2.slope.mtx
  --fsgd g1v1.fsgd \
  --C intercept.mtx \
  --C slope.mtx

Back to FSGD Examples

Two Groups (One Factor/Two Levels), One Covariate

This models the input as a single line (ie, an intercept and a slope). factor (eg, Age).

FSGD File (g1v1.fsgd)

GroupDescriptorFile 1
Title OSGM
Class Main
Variable Age
Input subject1 Main 30
Input subject2 Main 40

Nclasses = 1
Nvariables = 1

Regressors

Nregressors = Nclasses*(Nvariables+1) = 1*(1+1) = 2
Regressor1: All ones. Codes intercept/mean for Main
Regressor2: age for each subject. Codes age slope for Group 1

Contrasts

The number of columns in each contrast matrix must be the same as the number of regressors (Nregressors). If there is only one row in the contrast matrix, then the result will be a t-test and will have a sign. Reversing the signs in the contrast matrix will only change the sign of the output, not its magnitude. If there is more than one row, the result will be an F-test and will be unsigned.

Contrast 1 (intercept.mtx)

Null Hypothesis: is the intercept equal to 0?

1 0

This is a t-test with the intercept>0 being positive (red/yellow).

Contrast 2 (slope.mtx)

Null Hypothesis: is the slope equal to 0?

0 1

This is a t-test with the slope>0 being positive (red/yellow).

mri_glmfit command

This is an example invocation of mri_glmfit. Depending upon your application, you may have other options as well.

mri_glmfit \
  --glmdir g1v1 \
  --y y.mgh \
  --fsgd g1v1.fsgd \
  --C intercept.mtx \
  --C slope.mtx

Fsgdf2G1V (last edited 2022-01-04 14:27:33 by DougGreve)