QC Tab Guide

The Quality Control (QC) tab provides diagnostic visualizations to assess the quality of your multi-echo fMRI data and tedana processing.

Overview

The QC tab displays:

Brain maps derived from multi-echo fitting
Histograms of parameter distributions
Diagnostic time series plots

These visualizations help you identify data quality issues that may affect component classification.

Brain Maps

T2* Map

What it shows: The estimated T2* relaxation time at each voxel.

How to interpret:

Region	Expected T2*	Notes
Gray matter	30-50 ms	Varies by field strength
White matter	40-60 ms	Slightly longer than GM
CSF	Very long	May appear saturated
Air/bone	Very short/zero	Signal dropout

What to look for:

Uniform values in gray matter: Good data quality
Spotty or noisy: Motion or acquisition issues
Very low values near sinuses: Expected susceptibility dropout
Asymmetric patterns: Potential shimming issues

Field Strength Matters

T2* values scale with field strength. Values above are approximate for 3T.

S0 Map

What it shows: The estimated signal intensity at TE=0 (proton density weighted).

How to interpret:

Feature	Interpretation
Bright gray matter	Normal
Bright CSF	Normal
Dark white matter	Normal
Very bright spots	Possible vessels or artifacts
Dark spots in brain	Signal dropout

What to look for:

Smooth intensity: Good coil sensitivity
Bright edges: Potential surface artifacts
Rings or bands: Acquisition artifacts

RMSE Map

What it shows: Root Mean Square Error of the multi-echo fitting.

How to interpret:

RMSE Level	Interpretation
Low (dark)	Good model fit
High (bright)	Poor fit, unreliable estimates

What to look for:

Low RMSE in gray matter: Model fits well
High RMSE at edges: Expected due to partial voluming
High RMSE in brain interior: Potential issues with:
- Motion
- Physiological noise
- Multi-echo timing errors

Histograms

Histograms show the distribution of values across the brain.

T2* Histogram

X-axis: T2* values (ms)
Y-axis: Number of voxels

What to look for:

Main peak at 30-50 ms: Normal gray matter
Secondary peak: May indicate white matter or different tissue
Long tail to high values: CSF contribution
Spike at zero: Masked voxels

S0 Histogram

Shows distribution of signal intensity
Should show smooth distribution
Outliers may indicate artifacts

Time Series Plots

Diagnostic temporal plots help identify:

Global signal fluctuations: Large-scale intensity changes
Motion effects: Sudden shifts in signal
Scanner drift: Slow intensity changes over time

Using QC for Classification

QC information can inform your component classification:

QC Finding	Classification Impact
High RMSE region	Components localized here may be unreliable
Signal dropout	Expect fewer reliable components in dropout areas
Motion artifacts in QC	Be more skeptical of edge-localized components

Troubleshooting

Maps Not Showing

If brain maps don't appear:

Check that files exist in your tedana output:
- T2starmap.nii*
- S0map.nii*
- rmse_statmap.nii*
These files are optional tedana outputs - not all tedana runs generate them

Unexpected Values

Issue	Possible Cause
T2* values too high	Wrong TE values provided
T2* values too low	Field inhomogeneity
Noisy S0 map	Motion during acquisition
High RMSE everywhere	Fundamental data quality issue

Best Practices

Review QC before classifying: Understanding data quality helps calibrate expectations
Note problem regions: If QC shows issues in specific areas, be cautious about components localized there
Compare across subjects: Consistent QC issues may indicate systematic problems
Document findings: Note any QC concerns for your records

QC Checklist

Before trusting your classifications:

[ ] T2* map shows reasonable values in gray matter
[ ] S0 map is smooth without major artifacts
[ ] RMSE is low in brain interior
[ ] No unexpected patterns in any maps
[ ] Histograms show expected distributions