Struct MediParams 

pub struct MediParams {

    pub lambda: f64,
    pub merit: bool,
    pub smv: bool,
    pub smv_radius: f64,
    pub smv_shrink_mask: bool,
    pub msmv: Option<MsmvParams>,
    pub data_weighting: i32,
    pub percentage: f64,
    pub cg_tol: f64,
    pub cg_max_iter: usize,
    pub max_iter: usize,
    pub tol: f64,
    pub b0: f64,
    pub echo_times: Vec<f64>,
    pub te: EchoTime,
    pub csf: bool,
    pub lambda_csf: f64,
    pub csf_r2s_threshold: f64,
    pub csf_erode: bool,
    pub zeropad: (usize, usize, usize),
}

MEDI parameters (MEDI_L1.m / parse_QSM_input.m defaults)

Fields

lambda: f64

Weight of the data-fidelity term ('lambda'): larger fits the data more closely

merit: bool

MERIT: reweight the noise by the data residual after each iteration ('merit')

smv: bool

MEDI-SMV ('smv'; false is 'nosmv'): high-pass the local field with the spherical mean value kernel, RDF − SMV(RDF), and fit it with the SMV-filtered dipole kernel (1 − SphereK)·D and the SMV-propagated noise sqrt(SMV(N²) + N²). This removes the residual background a background removal leaves near the edge. It is applied to the local field, not the total field (see the module docs). On by default, as in MEDI_L1.m.

smv_radius: f64

SMV radius in mm

smv_shrink_mask: bool

With SMV, erode the mask to where the SMV of the mask exceeds 0.999 (smv_shrink_mask, default on). With it off, the field is still high-passed, the kernel filtered and the noise propagated, on the full mask, as the 2024.11.26 MEDI_L1.m does. The mSMV prerelease’s MEDI_L1.m skips the field high-pass in this case and keeps the rest; medi follows the release (table in the module docs). The edge band within the SMV radius then holds RDF − SMV(RDF) of a sphere that reaches outside the mask. Ignored without SMV, and with msmv, which keeps the mask, as MEDI_L1 does.

msmv: Option<MsmvParams>

MEDI-mSMV (MEDI_L1(..., 'msmv', smv_radius)): replace the SMV step’s mask erosion and field high-pass with msmv on the full mask, keeping the SMV-filtered dipole kernel and noise. Takes the local field from a background removal (the authors used PDF, V-SHARP and LBV) on its mask; implies SMV (radius smv_radius). Needs at least three echoes: the vessel mask uses R2* from arlo.m on the echo magnitudes, as MEDI’s pipeline computes it. None (default) is plain MEDI or MEDI-SMV. See the module docs.

data_weighting: i32

Data weighting: 0 uniform (1 everywhere), 1 SNR (mask / n_std, mean 1 in the mask)

percentage: f64

Edge fraction for the gradient mask ('percentage'): the threshold is set so that the number of gradient components (all three directions) above it is this fraction of the number of voxels in the mask

cg_tol: f64

CG relative residual tolerance

cg_max_iter: usize

Maximum CG iterations per Gauss-Newton step

max_iter: usize

Maximum Gauss-Newton iterations

tol: f64

Stop when ||dχ|| / ||χ|| falls below this

b0: f64

Field strength in Tesla (scan parameter, for the radian conversion)

echo_times: Vec<f64>

Echo times in seconds, one per echo magnitude (scan parameter). The radian conversion picks one with te; the CSF mask uses R2* from all of them (arlo.m).

te: EchoTime

Which echo time the radian conversion uses. Default EchoTime::Spacing, MEDI’s delta_TE: MEDI_L1.m works on the field in radians over delta_TE (x/(2*pi*delta_TE*CF)*1e6), which its readers set to TE(2) − TE(1), or TE for one echo (Read_Siemens_DICOM.m lines 176-178).

csf: bool

MEDI+0: zero-reference to the ventricular CSF (extract_CSF.m on arlo.m R2*, and MEDI_L1’s lambda_CSF term). On by default, as in MEDI’s pipeline. Needs at least three echoes; with fewer, MEDI runs without it (see the module docs).

lambda_csf: f64

Weight of the CSF term ('lambda_CSF')

csf_r2s_threshold: f64

extract_CSF.m’s thresh_R2s: voxels below this R2* (Hz) are CSF candidates

csf_erode: bool

extract_CSF.m’s flag_erode: erode the mask by a 10 mm SMV before finding the CSF

zeropad: (usize, usize, usize)

Voxels of zeros appended along x, y and z before the inversion ('zeropad'). With SMV, axes that are still odd get one more (see the module docs).

Trait Implementations

impl Clone for MediParams

fn clone(&self) -> MediParams

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for MediParams

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for MediParams

fn default() -> Self

Returns the “default value” for a type.

impl Serialize for MediParams

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.