MRI k-space

3 ways to acquire data

1. sequential- all data from slice one, all data from slice 2 2. 2D volumetric- fills one line from slice one, one line from slice 2, next from 1, next from 2 and so on 3. 3D volumetric- data is acquired from entire volume of tissue rather than separate slices ^ 3D uses slice encoding; at the end of the acquisition the volume is separated into slices

What is k-space?

A spatial frequency domain, where information is stored about the frequency of a signal and where it comes from. (where data is acquired and stored)

Blipping

A type of gradient switching where the amplitude gradually increases until it reaches maximum negativity polarity, and the bottom line of k-space is filled

Reduction/Acceleration factor

As 4 lines are acquired per TR, the scan time decreases by a factor of 4. The reduction factor equals the number of channels in the configuration

Transmit bandwidth

Bandwidth used in RF excitation pulse to excite a slice. Permits slices of a certain thickness.

Simplest method of k-space filling

Data is stored in horizontal lines that are parallel to the frequency and perpendicular to the phase

Data points

Determined by frequency matrix, frequencies in the echo are digitalized, and data is acquired and located in line in k-space. They are acquired over time, and are laid out in k-space during the scan and mathematically converted into information related to amplitude via FFT

Image matrix

Determined by the number of data points in k-space The phase matrix depends on how many lines of k-space are filled with data points It is therefore the total number of data points in each column of k-space The frequency matrix is the number of data points in each row of k-space

Slice-select gradient

Determines which area of k-space is being transversed

Central portion of k-space

High signal amplitude, low resolution

Frequency axis

Horizontal and centered

Nyquist theorem

How the most optimum digital sampling frequency is determined. Calculates the minimum digital sampling frequency needed to acquire enough data points to create an accurate image. States that when digitizing a modulation of several frequencies, the highest frequency present in the modulation must be sampled at least twice as frequently to accurately digitize it. Used to avoid aliasing artifact

Outer portion of k-space

Low signal amplitude, high resolution

Frequency/Amplitude modulation

Multiple waveforms in the echo are simplified into one waveform that represents the average amplitude of all the different frequencies in the echo at different time points. This is also the echo that occurs at TE. Frequency encoding=frequency matrix=data points in each column of k-space

Chest Drawers analogy

Number of lines filled is the number drawers. Drawers = phase matrix, socks = data points =frequency matrix Each gradient determines how and when the drawers are filled

Receive Bandwidth

Parameter that has the same numerical values as the digital sampling frequency (when nyquist is used) The range of frequencies accurately sampled during the sampling window. Determined by applying a filter to the frequencies encoding by the frequency encoding gradient

Phase-encoding gradient filling

Phase-encoding gradient picks which line of k-space in a certain TR period. Both the polarity and the slope of the phase gradient are altered every TR to produce multiple pixels in the phase direction of the image. Positive polarity phase-encoding gradients pick lines in the top half of the gradient, negative picks bottom half. Steep gradients pick outer lines, shallow picks central. Steep produces more phase shift Phase encoding=phase matrix=data points in row of k-space

Digital sampling rate/frequency

Rate at which sampling occurs

Phase FOV dimension is changed by

Rectangular/asymmetric FOV Antialiasing Parallel imaging

Frequency resolution

Resolution in k-space

Parallel imaging (for resolution)

Resolution: more SNR: less Scan time: same Purpose: increases resolution

Rectangular FOV (phase short axis)

Resolution: same SNR: less Scan time: less Purpose: Reduces scan time when anatomy is rectangular

Partial averaging

Resolution: same SNR: less Scan time: less Purpose: Reduces time when SNR is good

Antialiasing (Siemens/newer phillips)

Resolution: same SNR: more Scan time: more Purpose: eliminates antialiasing

Turbo-spin echo (for scan time)

Resolution: same SNR: same Scan time: less Purpose: reduces scan time

Partial echo

Resolution: same SNR: same Scan time: same Purpose: Automatic for a short TE

Antialiasing (GE/older phillips)

Resolution: same SNR: same Scan time: same Purpose: eliminates antialiasing

Respiratory compensation

Resolution: same SNR: same Scan time: same Purpose: reduces respiratory artifact

Parallel imaging (for scan time)

Resolution: same SNR: less Scan time: less Purpose: reduces scan time Fills multiple lines of k-space per TR

RF excitation pulses

Result in the system centering itself in the center of k-space

As phase matrix increases

Scan time also increases

Cartesian filling

Simplest method of filling. K-space is filled in a linear manner from top to bottom or bottom to top

Spatial resolution

Size of each pixel

Acquisition window

The duration of the frequency encoding gradient, also called sampling time or sampling window. The system measures the echo at certain points. Every time this measurement is taken it is stored as a data point in k-space.

Readout/Measurement gradient

The frequency encoding gradient is switched on while system reads the echo and digitizes it

Analog-to-Digital conversion (ADC)

The process of digitizing the above modulation (echo). Done either inside the receiver could assembly of the body pf the MRI scanner.

Conjegate symmety

The symmetry of K-space Used to reduce scan times

Sampling interval

Time interval between each data point Calculated by dividing the digital sampling frequency by 1 If the digital sampling frequency increases, sampling interval decreases

Phase axis

Vertical and centered perpendicular to frequency axis

single-shot (SS) imaging

Where all of the lines are filled at once. collected all from one single echo train No TR bc there is only 1 RF excitation pulse

FFT (Fast Fourier Transform)

a numerical technique for determining the spectrum of a digital sound. Required to "unlock" each data point and calculate the signal intensity for each pixel position in the slice

Pseudo-frequency

frequency that is indirectly derived from a change of phase. Each pixel has it's own pseudo-frequency, so hundreds are obtained throughout the scan.

Scan time

multiply: TR Phase matrix NSA (number of signal averages)

RF rephrasing pulses

result in a point in k-space flipped to the mirror point on the opposite side of k-space

Phase axis & FOV

the difference in phase shift between each data point in each column is inversely proportional to the size of the phase FOV