bartorch.tools.mobafit

Contents

bartorch.tools.mobafit#

bartorch.tools.mobafit(enc, echo_contrast_images, *, B=False, M=False, I=False, L=False, G=False, D=False, S=False, m=None, a=False, i=None, g=False, init=None, scale=None, levenberg_marquardt=False, reg_against_init=False, liniter=None, min_flag=None, max_flag=None, max_mag_flag=None, min=None, max=None, b1map=None, b0map=None, seq=None, sim=None, other=None, pool=None, cest=None, **extra)#

Pixel-wise fitting of physical signal models.

Runs bart mobafit.

Parameters:

  • enc (torch.Tensor) – Input array.

  • echo_contrast_images (torch.Tensor) – Input array.

  • B (bool) – bSSFP (-B)

  • M (bool) – MOLLI (-M)

  • I (bool) – Inversion Recovery: f(M0, R1, c) = M0 * (1 - exp(-t * R1 + c)) (-I)

  • L (bool) – Inversion Recovery Look-Locker: f(Mss, M0, R1s) = Mss - (Mss + M0) * exp(-t * R1s) (-L)

  • G (bool) – MGRE (-G)

  • D (bool) – diffusion (-D)

  • S (bool) – Simulation based fitting (-S)

  • m (int | None) – Select the MGRE model from enum { WF = 0, WFR2S, WF2R2S, R2S, PHASEDIFF } [default: WFR2S] (-m)

  • a (bool) – fit magnitude of signal model to data (-a)

  • i (int | None) – Number of IRGNM steps (-i)

  • M – Multi-Pool-Lorentzian - Number of pools (-M)

  • g (bool) – use gpu (-g)

  • B – temporal (or other) basis (-B)

  • init (tuple[float, ...] | None) – Initial values of parameters in model-based reconstruction (--init)

  • scale (tuple[float, ...] | None) – Scaling (--scale)

  • levenberg_marquardt (bool) – Use Levenberg-Marquardt instead of Gauss-Newton (--levenberg-marquardt)

  • reg_against_init (bool) – Use initial guess as regularization target with IRGN (--reg-against-init)

  • liniter (int | None) – (iterations for solving linearized problem) (--liniter)

  • min_flag (int | tuple[int, ...] | None) – Parameter maps with a minimum constraint, by index. (--min-flag)

  • max_flag (int | tuple[int, ...] | None) – Parameter maps with a maximum constraint, by index. (--max-flag)

  • max_mag_flag (int | tuple[int, ...] | None) – Parameter maps with a maximum magnitude constraint, by index. (--max-mag-flag)

  • min (tuple[float, ...] | None) – Min bound (map must be selected with “min-flag”) (--min)

  • max (tuple[float, ...] | None) – Max bound (map must be selected with “max-flag” or “max-mag-flag”) (--max)

  • b1map (torch.Tensor | None) – Input B1 map as cfl file (--b1map)

  • b0map (torch.Tensor | None) – Input B0 map as cfl file (--b0map)

  • seq (str | None) – configure sequence parameters for simulation based fitting (--seq)

  • sim (str | None) – configure simulation parameters (--sim)

  • other (str | None) – configure other simulation parameters (--other)

  • pool (str | None) – configure pool parameters for BMC simulation (--pool)

  • cest (str | None) – configure parameters for CEST simulation (--cest)

  • **extra (Any) – Further BART flags, passed through by name.

Returns:

coefficients

Return type:

torch.Tensor

Contents