2014 M.S. Cohen all rights reserved

Size: px

Start display at page:

Edmund Johnson
5 years ago
Views:

2 IMAGE QUALITY / ARTIFACTS

3 SYRINGOMYELIA Source Surgery is usually recommended for syringomyelia patients. The main goal of surgery is to provide more space for the cerebellum (Chiari malformation) at the base of the skull and upper neck, without entering the brain or spinal cord. This results in flattening or disappearance of the primary cavity. If a tumor is causing syringomyelia, removal of the tumor is the treatment of choice and almost always eliminates the syrinx.

4 Truncating the Fourier Data Results in Distortions (edge ringing) of High Spatial Frequencies

5 This is equivalent to sampling only a portion of the raw data.

6 TRUNCATED RAW DATA Actual Object MR Image

7 THE SYRINGOMYELIA EPIDEMIC M. Cohen & D. Baird. Perspective on Science 7:231, 1999

8 TRUNCATION IN FOURIER DOMAIN Original Sample Frequencyencoded Signal -1 Apparent Signal F (s) Truncated Series

9 WHAT IS THE ACTUAL RESOLUTION OF MRI? Original Data MR Image Single pixel activation

THE ACTUAL RESOLUTION OF fmri http://ccn.ucla.

10 THE ACTUAL RESOLUTION OF fmri

11 AN EQUATION IN RESOLUTION Because MR is an emission modality the temporal resolution, spatial resolution and contrast are inter-dependent: Signal = kb 0 (voxel size) imaging time contrast where B 0 is the field strength.

12 CONTRAST TO NOISE RATIO

13 CNR VS. RESOLUTION Noise free Imaging time = 1X Imaging time = 2X Imaging time = 4X 64 X X 128 Minimum Imaging Time 256 X 256

14 CNR VS. RESOLUTION 256 X 256 Signal/Noise Ratio Held Constant 128 X X 64 Imaging time = 16X Imaging time = 4X Imaging time = 1X

15 CNR VS. RESOLUTION Noise free 16 averages 4 averages 1 average 64 X X X 256 Imaging Time Held Constant

16 BANDWIDTH AND READOUT Position is encoded by FREQUENCY Bandwidth refers to the Frequency Difference from the center of the image to its edge: Frequency per pixel = 2* Bandwidth = number of pixels Bandwidth decreases with readout duration: Bandwidth = number of pixels 2 * readout duration 1 readout duration

17 BANDWIDTH AND SNR Decreasing the Bandwidth Improves SNR: Imaging Time is INCREASED and high frequency noise is excluded Signal Intensity Narrow Wide Bandwidth Frequency Noise

18 BANDWIDTH BW=4kHz BW=8kHz BW=16kHz TE=11-14 NEX=1 Thick=3mm TR=500 Matrix=256x256

19 BANDWIDTH BW=4kHz BW=8kHz BW=16kHz TE=11-14 NEX=1 Thick=3mm TR=500 Matrix=256x256?

20 THE ORIGIN OF CHEMICAL SHIFT In water, electrons move from Hydrogen towards Oxygen. Electrons in lipid are shared equally between Hydrogen and Oxygen This exposes the Proton to a slightly higher magnetic field. Water Lipid Resonance Frequencies Higher Frequency

21 CHEMICAL SHIFT ARTIFACT Higher Frequency If the frequency width of each pixel is less than the frequency difference between water and lipid, then water and lipid will appear in separate pixels

22 CHEMICAL SHIFT The Fat-Water chemical shift is about 3.5 ppm or: Which is: with a 32 khz readout Tesla < 1 pixel Tesla 1 pixel Tesla > 1 pixel Tesla 3.5 pixels d Amplitude Water Fat frequency Lowering the Bandwidth/pixel increases the Chemical Shift in pixels

23 SHAPE AND BANDWIDTH

24 DISTORTION INCREASES WITH FIELD STRENGTH Variation in sample magnetization of is proportional to field strength. High Field images lose more signal from field inhomogeneity

25 BLURRING FROM T2* DECAY T2* = 80 ms T2* = 10 ms

26 APODIZATION FROM LONG READOUTS Phantom Readout = 2T2* Readout = 4T2* Exercise in Afternoon Lab

27 EPI READOUT DURATIONS 1 MR Signal UCLA 64x GE Product 64x128 UCLA 128x128 GE Product 128x128 Stanford Spiral 128x128 0 T2* signal decay (T2* ~ 45 msec)

28 MOTION ARTIFACT Exercise in Afternoon Lab

29 GRADIENTS AND PHASE + 0 T/2 T Phase with respect to center

30 GRADIENTS AND PHASE x(t) = x 0 + vt + at ϕ = γ T 0 G(t)x(t)dt T = γ G(t) x + vt + at dt 0 = γ vt aT T/2 T

31 MOTION ARTIFACT Exercise in Afternoon Lab

32 ϕ = γ vt aT

33 ALIASING If the Sampling Rate is Less Than Twice the Signal Frequency, the Apparent Frequency is Ambiguous

34 SATURATION Exercise in Afternoon Lab

35 SATURATION (CLIPPING) Amplifier Limit

36 SIGNAL INHOMOGENEITY Gradient Echo spgr, tr/te/flip 60/6/20 Spin Echo tr/te 500/11 36

37 7 TESLA

38 RF PENETRATION Higher RF Field & Effective Flip Angle Lower The well-known skin effect results in greater current density (and flip angle) on the surface of conducting objects than at their center

40 K-SPACE TRAVERSALS tr ms Spiral EPI 4-16 ms Conventional ms Rectilinear EPI ms 40

41 SPIRAL SCANS (128X128) 41

3 TESLA ECHO PLANAR IMAGES 128X256 10 of 20 4 mm

42 3 TESLA ECHO PLANAR IMAGES 128X of 20 4 mm sections, each acquired in 78 ms. TE = 55 msec, 1.5 mm pixels 42

43 ? Exercise in Afternoon Lab

46 Stimulated Echo

50 ?

51 Quadrature Ghost

52 ? Eddy Currents

53 ? RF noise

54 ? Parasitic conduction

55 cochlear implant Metal artifact

56 beeswax Foreign bodies

57 Metal artifact

58 EPI ghost

59 ?

60 ?

61 ?

62 ?

63 ?

64 ?

65 ?

67 ?

68 ?

70 ?

71 ?

75 INSTRUMENT VARIATION 1. System Instability Mean Intensity Variation

76 THE WEISSKOFF PLOT The Expected Standard Deviation of the Mean Signal of a Region over Time Falls with the Square Root of the Number of Voxels L Coefficient of Variation ROI Edge Length

77 THE WEISSKOFF PLOT 0.2 Coefficient of Variation 0.1 Measured Theoretical Deviations from the Theoretical Curve are Evidence of Correlated Noise RDC = ROI Edge Length RDC (Radius of Decorrelation) is a Single Point Quantification of the Weisskoff Plot

78 WEISSKOFF PLOT ROI Length ROI Length Weisskoff R. Magn Reson Med 36:643 Friedman and Glover, JMRI 23:827

79 SCANNER COMPARISONS Friedman and Glover, JMRI 23:827

80 GLOBAL MEAN SCALING - OFF

81 GLOBAL MEAN SCALING - ON

82 INTERPOLATION This location was not acquired Native Resolution Bilinear Interpolation

83 DRIFT

84 INSTRUMENT VARIATION 2. The mystery of scanner drift.

85 CHARACTERIZE YOUR TOOLS Test Statistics are Effect/Variance Variance includes: Intrasubject (motion, attention, physiology, fatigue, ) Intersubject variance (position, morphology, performance, pathology, physiology, ) Experimental Variance (uncontrolled variables, stimulation variance, ) Instrument Variance Sitewise Variance True Random Noise

86 Basic MRI Artifacts Motion Shape and bandwidth Aliasing Time Series Artifacts Data Analysis ARTIFACTS IN FMRI Signal Drifts Field Changes RF Penetration/Uniformity Signal Voids Data Spikes Global Normalization Motion Smoothing False Positives and False Negatives Excess variance Motion Cardiac Pulsation Respiration Timing Error Over-aggressive Smoothing Apodization Chemical Shift K-Space errors (spiral) Ratty Image Quality Wrong Statistical Test

Image Quality/Artifacts Frequency (MHz)

Image Quality/Artifacts Frequency (MHz) The Larmor Relation 84 Image Quality/Artifacts (MHz) 42 ω = γ X B = 2πf 84 0.0 1.0 2.0 Magnetic Field (Tesla) 1 A 1D Image Magnetic Field Gradients Magnet Field Strength Field Strength / Gradient Coil