statistical analysis i. basic hypothesis-driven analyses

41
Statistical Analysis I. Basic Hypothesis- Driven Analyses fMRI Graduate Course November 13, 2002

Upload: talmai

Post on 09-Jan-2016

55 views

Category:

Documents


0 download

DESCRIPTION

Statistical Analysis I. Basic Hypothesis-Driven Analyses. fMRI Graduate Course November 13, 2002. When do we not need statistical analysis?. Inter-ocular Trauma Test (Lockhead, personal communication). Why use statistical analyses?. Replaces simple subtractive methods - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Statistical Analysis I. Basic Hypothesis-

Driven Analyses fMRI Graduate Course

November 13, 2002

Page 2: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

When do we not need statistical analysis?

0.8

0.85

0.9

0.95

1

1.05

1.1

1.15

1.2

1 51 101 151 201 251

Inter-ocular Trauma Test (Lockhead, personal communication)

Page 3: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Why use statistical analyses?

• Replaces simple subtractive methods– Signal highly corrupted by noise

• Typical SNRs: 0.2 – 0.5

– Sources of noise• Thermal variation (unstructured)• Physiological variability (structured)

• Assesses quality of data– How reliable is an effect?– Allows distinction of weak, true effects from strong,

noisy effects

Page 4: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Statistical Parametric Maps

• 1. Brain maps of statistical quality of measurement– Examples: correlation, regression approaches– Displays likelihood that the effect observed is due to

chance factors– Typically expressed in probability (e.g., p < 0.001)

• 2. Effect size– Determined by comparing task-related variability and

non-task-related variability– Signal change divided by noise (SNR)– Typically expressed as t or z statistics

Page 5: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Why use effect size measures?

• Dissociate size of signal change from reliability of signal change– Understanding reliability of change allows quantification of error

probabilities

• Types of Errors– Type I: Rejecting null hypothesis when it is true

• Calling “active” voxels that really have no activity• To minimize false positives, adopt a high threshold for significance

– Type II: Accepting null hypothesis when it is false• Calling “inactive” voxels that are really associated with the task• To minimize incorrect rejections, adopt a low threshold for

significance

Page 6: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Simple Statistical Analyses

• Common– t-test across conditions– Fourier– t-test at time points – Correlation

• General Linear Model • Other tests

– Kolmogorov-Smirnov – Iterative Connectivity Mapping

Page 7: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

0.8

0.85

0.9

0.95

1

1.05

1.1

1.15

1.2

1 51 101 151 201 251

T – Tests across Conditions

• Compares difference between means to population variability– Uses t distribution– Defined as the likely distribution

due to chance between samples drawn from a single population

• Commonly used across conditions in blocked designs

• Potential problem: Multiple Comparisons

5%

Page 8: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Drift Artifact and T-Test

Page 9: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Fourier Analysis• Fourier transform: converts information in time domain to frequency

domain– Used to change a raw time course to a power spectrum– Hypothesis: any repetitive/blocked task should have power at the task

frequency

• BIAC function: FFTMR– Calculates frequency and phase plots for time series data.

• Equivalent to correlation in frequency domain– At short durations, like a sine wave (single frequency)– At long durations, like a trapezoid (multiple frequencies)

• Subset of multiple regression– Same as if used sine and cosine as regressors

Page 10: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

12s on, 12s off Frequency (Hz)

Pow

er

Page 11: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

-3%-2%

-1%0%1%

2%3%

1 12 23 34 45 56 67 78 89 100 111 122

Image

Left-Right

Right-Left

-3%-2%-1%0%1%2%3%

1 12 23 34 45 56 67 78 89 100 111 122

Image

Left-Right

Right-Left

0

400

800

1200

1600

0.005 0.047 0.089 0.130 0.172 0.214 0.255 0.297

Frequency

0

200

400

600

800

1000

1200

1400

0

20 40 60 80

100

120

140

160

180

200

220

240

260

280

300

320

340

Phase Angle (Degrees)

Spe

ctra

l Pow

er a

t 0.0

58 H

z

0

200

400

600

800

1000

1200

1400

0 30 60 90 120 150 180 210 240 270 300 330

Phase Angle (Degrees)

Spe

ctra

l Pow

er a

t 0.0

58 H

z

0

400

800

1200

1600

0.005 0.047 0.089 0.130 0.172 0.214 0.255 0.297

Frequency

Page 12: Statistical Analysis  I. Basic Hypothesis-Driven Analyses
Page 13: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

T/Z – Tests across Time Points

• Determines whether a single data point in an epoch is significantly different from baseline

• BIAC Tool: tstatprofile– Creates:

• Avg_V*.img• StdDev_V*.img• ZScore_V*.img

-20

-15

-10

-5

0

5

10

15

20

25

30

35

-5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13

Page 14: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Correlation

• Special case of General Linear Model– Blocked t-test is equivalent to correlation with square

wave function– Allows use of any reference waveform

• Correlation coefficient describes match between observation and expectation– Ranges from -1 to 1– Amplitude of response does not affect correlation

directly

• BIAC tool: tstatprofile

Page 15: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Problems with Correlation Approaches

• Limited by choice of HDR– Poorly chosen HDR can significantly impair power

• Examples from previous weeks

– May require different correlations across subjects

• Assume random variation around HDR– Do not model variability contributing to noise (e.g.,

scanner drift)• Such variability is usually removed in preprocessing steps

– Do not model interactions between successive events

Page 16: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Kolmogorov – Smirnov (KS) Test

• Statistical evaluation of differences in cumulative density function– Cf. t-test evaluates differences in mean

ns

p < 10-30

p < 10-30

Page 17: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Iterative Connectivity Mapping

• Acquire two data sets– 1: Defines regions of interest

and hypothetical connections– 2: Evaluates connectivity

based on low frequency correlations

• Use of Continuous Data Sets– Null Data– Task Data– Can see connections

between functional areas (e.g., between Broca’s and Wernicke’s Areas)

Hampson et al., Hum. Brain. Map., 2002

Page 18: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Use of Continuous Tasks to Evaluate Functional Connectivity

Hampson et al., Hum. Brain. Map., 2002

Page 19: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

The General Linear Model

Page 20: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Basic Concepts of the GLM

• GLM treats the data as a linear combination of model functions plus noise– Model functions have known shapes– Amplitude of functions are unknown– Assumes linearity of HDR; nonlinearities can be

modeled explicitly

• GLM analysis determines set of amplitude values that best account for data– Usual cost function: least-squares deviance of

residual after modeling (noise)

Page 21: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Signal, noise, and the General Linear Model

MYMeasured Data

Amplitude (solve for)

Design Model

Noise

Cf. Boynton et al., 1996

Page 22: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Form of the GLMD

ata=

N T

ime

Poi

nts

Model

N T

ime

Poi

nts

Model Functions

* Amplitudes

Model Functions

+

Noi

se

Page 23: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Implementation of GLM in SPM

Page 24: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

The Problem of Multiple Comparisons

Page 25: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

The Problem of Multiple Comparisons

P < 0.001 (32 voxels)P < 0.01 (364 voxels)P < 0.05 (1682 voxels)

Page 26: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Options for Multiple Comparisons

• Statistical Correction (e.g., Bonferroni)– Gaussian Field Theory

• Cluster Analyses

• ROI Approaches

Page 27: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Statistical Corrections

• If more than one test is made, then the collective alpha value is greater than the single-test alpha– That is, overall Type I error increases

• One option is to adjust the alpha value of the individual tests to maintain an overall alpha value at an acceptable level– This procedure controls for overall Type I error – Known as Bonferroni Correction

Page 28: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

0

0.2

0.4

0.6

0.8

1

1.2

Number of Comparisons

Pro

bab

ility

of

Typ

e I E

rro

r

0.00000001

0.0000001

0.000001

0.00001

0.0001

0.001

0.01

0.1

1

Co

rrec

ted

Alp

ha

Val

ue

Type I Probability

Adjusted Alpha

Page 29: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Bonferroni Correction

• Very severe correction– Results in very strict significance values for even

medium data sets– Typical brain may have about 15,000-20,000

functional voxels• PType1 ~ 1.0 ; Corrected alpha ~ 0.000003

• Greatly increases Type II error rate• Is not appropriate for correlated data

– If data set contains correlated data points, then the effective number of statistical tests may be greatly reduced

– Most fMRI data has significant correlation

Page 30: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Gaussian Field Theory

• Approach developed by Worsley and colleagues to account for multiple comparisons– Forms basis for much of SPM

• Provides false positive rate for fMRI data based upon the smoothness of the data– If data are very smooth, then the chance of

noise points passing threshold is reduced

Page 31: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Figures from http://www.irsl.org/~fet/Presentations/wavestatfield/wavestatfield.html

Page 32: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Cluster Analyses

• Assumptions– Assumption I: Areas of true fMRI activity will typically

extend over multiple voxels– Assumption II: The probability of observing an

activation of a given voxel extent can be calculated

• Cluster size thresholds can be used to reject false positive activity– Forman et al., Mag. Res. Med. (1995)– Xiong et al., Hum. Brain Map. (1995)

Page 33: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

How many foci of activation?

Data from motor/visual event-related task (used in laboratory)

Page 34: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

How large should clusters be?

• At typical alpha values, even small cluster sizes provide good correction– Spatially Uncorrelated Voxels

• At alpha = 0.001, cluster size 3 reduces Type 1 rate to << 0.00001 per voxel

– Highly correlated Voxels• Smoothing (FW = 0.5 voxels) increases needed cluster size

to 7 or more voxels

• Efficacy of cluster analysis depends upon shape and size of fMRI activity– Not as effective for non-convex regions– Power drops off rapidly if cluster size > activation size

Data from Forman et al., 1995

Page 35: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

ROI Comparisons

• Changes basis of statistical tests – Voxels: ~16,000– ROIs : ~ 1 – 100

• Each ROI can be thought of as a very large volume element (e.g., voxel)– Anatomically-based ROIs do not introduce bias

• Potential problems with using functional ROIs– Functional ROIs result from statistical tests– Therefore, they cannot be used (in themselves) to

reduce the number of comparisons

Page 36: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Are there differences between voxel-wise and ROI analyses?

Page 37: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Summary of Multiple Comparison Correction

• Basic statistical corrections are often too severe for fMRI data

• What are the relative consequences of different error types? – Correction decreases Type I rate: false positives– Correction increases Type II rate: misses

• Alternate approaches may be more appropriate for fMRI– Cluster analyses– Region of interest approaches– Smoothing and Gaussian Field Theory

Page 38: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Fixed and Random Effects Comparisons

Page 39: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

How do we compare across subjects?

• Fixed-effects Model– Uses data from all subjects to construct statistical test– Examples

• Averaging across subjects before a t-test• Taking all subjects’ data and then doing an ANOVA

– Allows inference to subject sample

• Random-effects Model– Accounts for inter-subject variance in analyses– Allows inferences to population from which subjects are drawn– Especially important for group comparisons– Beginning to be required by reviewers/journals

Page 40: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

How are random-effects models run?

• Assumes that activation parameters may vary across subjects– Since subjects are randomly chosen, activation parameters may

vary within group– Fixed-effects models assume that parameters are constant

across individuals

• Calculates descriptive statistic for each subject – i.e., t-test for each subject based on correlation

• Uses all subjects’ statistics in a one-sample t-test – i.e., another t-test based only on significance maps

Page 41: Statistical Analysis  I. Basic Hypothesis-Driven Analyses

Summary of Hypothesis Tests

• Simple experimental designs– Blocked: t-test, Fourier analysis– Event-related: correlation, t-test at time points

• Complex experimental designs– Regression approaches (GLM)

• Critical problem: Minimization of Type I Error– Strict Bonferroni correction is too severe– Cluster analyses improve– Accounting for smoothness of data also helps

• Use random-effects analyses to allow generalization to the population