Title: Artifacts in Cardiac MRI
1Artifacts in Cardiac MRI
- Andreas Sigfridsson, IMV/IMT/CMIV
ltsigge_at_imv.liu.segt - 2007-04-27
2Dynamic imaging
- Images are acquired line-by-line in k-space
- Motion during acquisition causes artifacts, not
blurring - Special care must be taken
- Triggering
- Acquisition ordering
3Example
- Simulation example
- 256x256, TR20 ms, total scan time 5 s
Image domain
k-space
4Signal saturation
- Signal recovery - relaxation
5Signal saturation
6Varying signal during inflow
- The MR signal is increased during inflow
- Signal increase related to flow velocity
- More magnetization because of T1 decay/recovery
Flowing spins, fresh signal
Static spins, saturated signal
7Varying signal intensity
- Simulated six sinusoidal heart beats
Intensity
Image (animation)
8Varying signal intensity, results
High contrast
Normal contrast
9Varying signal intensity, k-space
- Image artifacts because of k-space sampling
- Different lines are acquired with different
signal intensity
10Varying signal intensity, analysis
11Slow sine variation
12Varying signal intensity, results
High contrast
Normal contrast
13Fast sine variation
Intensity
FT(Intensity)
14Fast sine variation
Intensity
FT(Intensity)
15Physiological variation
- Aortic flow (handcrafted, not actual data)
Intensity (whole scan)
Intensity (one beat)
16Physiological variation
Intensity variation (animation)
17Physiological variation
FT(Intensity)
Intensity
18Physiological variation
High contrast
Normal contrast
19Remedies
- Triggering
- Wait for ECG R-tag
- Wait additional trigger delay time
- Acquire one (or several) k-space lines
- Acquisition reordering
- Measure current cardiac (or respiratory phase)
- Choose k-space line to acquire
- Creates a slow varying intensity curve along
k-space
20Triggering - random phase
- If triggering is not accurate, random variations
will still be left
Intensity
FT(Intensity)
21Random phase, reconstructed
Normal contrast
High contrast
22Acquisition reordering
Intensity
FT(Intensity)
23Acquisition reordering
Ordered
Random
24Acquisition reordering
High contrast
Normal contrast
25Actual examples
26Dynamic imaging
- Images are acquired line-by-line in k-space
- Motion during acquisition causes artifacts, not
blurring - Special care must be taken
- Triggering
- Acquisition ordering
- Artifacts might be predictable through use of
Fourier transform