Utilities

A collection of helper functions that may be useful for testing code, building examples, or to construct new models.

Real data

fetch(dataset)

Fetch a dataset included in dcmri

Synthetic data

fake_aif([tacq, dt, BAT, field_strength, ...])	Synthetic AIF signal with noise.
fake_brain([n, tacq, dt, BAT, Tav, ...])	Synthetic brain images data generated using Parker's AIF, the Shepp-Logan phantom and a two-compartment exchange tissue.
fake_tissue([tacq, dt, BAT, Fb, vb, PS, vi, ...])	Synthetic data generated using Parker's AIF and a two-compartment exchange tissue.
fake_liver([tacq, dt, BAT, Tg, H, ve, Fp, ...])	Synthetic data for liver tissue.
fake_kidney([tacq, dt, BAT, Fp, Eg, fc, Tg, ...])	Synthetic data generated using Parker's AIF, and a multicompartment kidney model.
fake_tissue2scan([tacq, tbreak, dt, BAT, ...])	Synthetic data generated using Parker's AIF, a two-compartment exchange tissue and a steady-state sequence.

Synthetic images

shepp_logan(*params[, n, B0])

Modified Shepp-Logan phantom mimicking an axial slice through the brain.

Input functions

aif_parker(t[, BAT])	Population AIF model as defined by Parker et al (2006)
aif_tristan(t[, agent, dose, rate, BAT, ...])	Arterial input function with default parameters for young healthy volunteers.
aif_tristan_rat(t[, BAT, duration])	Population AIF model for rats measured with a standard dose of gadoxetate.
ca_injection(t, weight, conc, dose, rate, t0)	Contrast agent flux (mmol/sec) generated by step injection.

Useful constants

ca_conc(agent)	Contrast agent concentration
ca_std_dose(agent)	Standard injection volume (dose) in mL per kg body weight.
relaxivity([field_strength, tissue, agent, type])	Contrast agent relaxivity values in units of Hz/M
T1([field_strength, tissue, Hct])	T1 value of selected tissue types.
T2([field_strength, tissue])	T2 value of selected tissue types.
PD([tissue])	Relative proton density (PD) value of selected tissue types.
perfusion([parameter, tissue])	perfusion parameters of selected tissue types.

Convolution

Convolution is an essential mathematical tool for solving linear and stationary compartment models. Explicit numerical convolution is slow, and dcmri includes apart from a generic convolution method also some faster and more accurate functions for use in special cases where one or both of the factors have a known form.

conv(f, h[, t, dt, solver])	Convolve two 1D-arrays.
expconv(f, T[, t, dt])	Convolve a 1D-array with a normalised exponential.
biexpconv(T1, T2, t)	Convolve two normalised exponentials analytically.
nexpconv(n, T, t)	Convolve n identical normalised exponentials analytically
stepconv(f, T, D[, t, dt])	Convolve a 1D-array with a normalised step function.

Helper functions

sample(t, tp, Sp[, dt])	Sample a signal at given time points.
add_noise(signal, sdev)	Add noise to an MRI magnitude signal.
interp(y, x[, pos, floor])	Interpolate uniformly sampled data.