MR Physics 2: Generating an Image Flashcards
At a basic level how to we generate an image?
We need to measure the 1H2O signal in 3D
Distinguish signal from 3 orthogonal axis
Use 3 magnetic field gradients: Z, X, Y
Why do MRI scanners need magnetic field gradients?
Magnetic field gradients are needed to encode the signal spatially
They produce a linear variation in magnetic field intensity in a direction in space
This variation in magnetic field intensity is added to the main magnetic field, which is far more powerful
How are different magnetic fields generated?
The variation is produced by pairs of coils, placed in each spatial direction (X, Y & Z)
How is the frequency measured?
By adding variation in field intensity to the main magnetic field (the field that is kept homogenous through the shim coils and ferromagnetic blocks in the bore)
Frequency is a function of what?
The gradient and the position of the nuclei
What is the MRI pulse sequence?
A programmed set of changing magnetic gradients - a seqeunce of events that control the scanner to control magnetisation
What axis does slice selection use?
The Z-axis gradient
Describe how we do slice selection.
Turn on a z-axis gradient to define frequency in space
Apply a one-dimensional, linear magnetic field gradient during the period that the RF pulse is applied
A 90° pulse applied in conjunction with a magnetic field gradient will rotate spins which are located in a slice or plane through the object.
Water magnetisation with Larmor frequency matching rf pulse are excited
What axis does frequency encoding use?
The X-axis gradient
Describe frequency encoding.
Frequency-encoding may be used to define location within a slice or between slices.
Done once slice selection has been done in the Z-axis
Apply another magnetic gradient but this time in the X-axis - this changes the Larmor frequencies of the nuclei in a gradient along the x-axis.
Each segment now returns a signal of a different frequency depending on its location along the slice – as they are of different frequencies, they eventually become of different phases.
Adding the signals together gives a large signal at the start, when they are still all in phase, but this signal drops off as the phases diverge.
This gradient is called the “read out” or “frequency encoding” gradient.
What axis does phase encoding use?
The Y-axis gradient
Describe phase encoding.
Applied in the Y-axis gradient & while it is applied it modifies the spin resonance frequencies, inducing dephasing of the spins, which persists after the gradient is interrupted
This results in all the protons precessing in the same frequency but in different phases
On receiving this signal, each row of protons will be slightly out of phase- their signals are more or less out of phase
What is Fourier transform in the process of generating an image?
The Fourier transform is a mathematical technique that allows an MR signal to be decomposed into a sum of sine waves of different frequencies, phases, and amplitudes.
It resolves the frequency- and phase-encoded MR signals that compose k-space
Converts amplitude as a function of time to amplitude as a function of frequency
What is a 2D spin warp sequence?
Basic method of image generation
Describe the stages of 2D spin warp sequence.
- Slice is selected (Z axis gradient)
- Readout gradient (X axis gradient) data is acquired for each repetition of the sequence
- Phase encode gradient (Y axis gradient) is applied at different strengths before data acquisition
- Data is stored in a matrix called k-space
What is K space?
The raw data matrix - signals acquired for x and y dimensions are stored in a 2D matrix called k-space
It is a matrix of FIDs that sample spatial frequency
Describe information stored in the centre of K space.
Center of k-space has maximum signal which contributes mostly to intensity
It contains low spatial frequency information, determining overall image contrast, brightness, and general shapes
High amplitude
Describe the information stored at the edge of K space.
Edge of k-space ghas the smallest signal but contributes to the detail
It contains high spatial frequency information (edges, details, sharp transitions)
Low amplitude
How can resolution of K-space be increased?
By acquiring more data points
What does increasing the size of the K-space matrix (number of pixels in the matrix) do?
Increases spatial resolution – smaller pixels/voxels means better detail
Decreases signal – there are fewer photons per voxel so the signal is less
Increases scan time – more voxels need to be acquired becuase more signals need to be created (only in the phase encoding direction as each voxel requires a new signal)
How does increasing the field of view affect MRI image quality?
larger field of view =larger area imaged but the matrix size remains the same and so, to fill up a larger area, the voxel becomes larger
Increases the signal – a larger voxel means more signal received per voxel
Lower resolution – the voxels become larger
Increased viewing area
How does slice thickness affect the quality of the MRI image?
Want to reduce the amount of space between each slice to prevent sections being missed but when slices overlap an area of cross-talk results which causes artefacts
Increased slice gap =
Less cross-talk
Increased coverage – slices placed further apart and, therefore, cover a larger area.
What are gradient echo sequences?
A gradient echo is a manipulation of the FID signal that begins by applying an external dephasing gradient field across the specimen or tissue.
This gradient causes a calibrated change in local magnetic fields and alters the resonance frequencies slightly across the specimen.
This results in accelerated dephasing and ‘squelching/ scrambling’ of the FID.
In step 2, the process is reversed.
A rephasing gradient is applied with the same strength but opposite polarity to the dephasing gradient, reversing/ undoing the phase scramble.
What is a benefit of gradient echo sequences?
They permit the use of shorter TR values and this produces faster image acquisition.
Why do we sample K-space?
Understanding k-space sampling offers important insight into the properties of a sequence e.g. info about signal-to-noise ratio, image distortion, resolution and contrast
K-space is sampled at regular points to capture array of spatial frequencies
What is the most common scheme to sample K-space?
Cartesian
What are other schemes used to sample K-space?
Radial
Spiral
Zigzag
What is compressed sensing of K-space?
The symmetry of k-space can be exploited to reduce imaging time
k-space lines are sampled semi-randomly. The centre of k-space is sampled at higher density
An algorithm then reconstructs the missing data
