Feature #927 fabber quantification new by jan-petr · Pull Request #1176 · ExploreASL/ExploreASL

jan-petr · 2022-08-26T20:13:15Z

Linked issue

Closes #927

How to test

I've tested it with BBB_VUmc data (Siemens PCASL 3D GRASE) both with 20 and 64ch coils

Comments

I merged quant_Fabber inside quant_Basil, with if statements for b.modules.asl.bMultiTE, but for the creation of the txt with the options to run FSL, since they were very different, I kept xASL_quant_Basil_Options and xASL_quant_Fabber_Options

the xASL_fsl_RunFSL stayed the same as now I'm using my personal directory with the copy of FSL plus Fabber. This path I add to the dataPar.json as:

jan-petr

Very good!! I found several smaller bugs (that might not work for other sequences) and edited some comments and headers. I have partly fixed them, so I added a comment just so that you understand why I did that and can verify this. And there are also a few comments that still need fixing.

Before merging, we should run UnitTest and TestDataSet to be sure that we haven't messed up the quantification for non-DEBBIE sequences.

Functions/Quantification/xASL_quant_Basil.m

Modules/SubModule_Population/xASL_wrp_GetROIstatistics.m

Modules/xASL_module_Population.m

Modules/SubModule_ASL/xASL_wrp_ResampleASL.m

jan-petr · 2022-09-07T18:48:11Z

One more question - can we close #1085 as this is already done here?

jan-petr · 2022-09-12T16:38:19Z

One more question - can we close #1085 as this is already done here?

Closing. But please check if Texch maps are correctly exported and if we can generate stats for it.

HenkMutsaerts

Can you please update/clean the issue, so it's easier to read? Now it still has text from Michael Stritt, and it says that the issue is halfway done.

Functions/Quantification/xASL_quant_Basil.m

HenkMutsaerts · 2022-10-11T17:12:44Z

Functions/Initialization/xASL_init_DefinePaths.m

            x.D.RawEPIdir           = fullfile(x.D.PopDir, 'Raw_EPI_Check');
            x.D.T1_ASLREGDIR        = fullfile(x.D.PopDir, 'T1_ASLReg');
            x.D.TTCheckDir          = fullfile(x.D.PopDir, 'ATT_Check');
+			x.D.TExchCheckDir       = fullfile(x.D.PopDir, 'TExch_Check');


Is Texch now defined as the abbreviation we should use? Do Amnah etc use the same abbreviation?

Yes - that was the name from Amnah. Am I right @BeatrizPadrela

yes she does

Functions/Initialization/xASL_init_FileSystem.m

Functions/Quantification/xASL_quant_Basil.m

Modules/SubModule_ASL/xASL_wrp_ResampleASL.m

Modules/SubModule_ASL/xASL_wrp_VisualQC_ASL.m

Modules/xASL_module_ASL.m

jan-petr

See answers to your commits (I wrote some answers while checking the code, so to make them appear, I have to submit a review :)).

jan-petr · 2022-10-11T20:33:23Z

Functions/Quantification/xASL_quant_Basil.m

-			LDs = LDs(2:end);
+            PLD_vector_size = length(unique(x.Q.Initial_PLD));
+			PLDs = PLDs(1:end-(PLD_vector_size/x.Q.TimeEncodedMatrixSize)); % Removing the two longer PLDs - 3.4 and 3.6
+			LDs = LDs(1:length(PLDs));


That second comment was indeed wrong for ordinary multi-PLD, we don't skip anything. I have removed the second comment - we only skip the first volume of a block for Time encoded.

Only the comment was wrong. We don't repeat anything. 063b8d1

Modules/SubModule_ASL/xASL_wrp_ResampleASL.m

Functions/Quantification/xASL_quant_Basil.m

jan-petr · 2022-10-11T20:53:46Z

Functions/Initialization/xASL_init_DefinePaths.m

            x.D.RawEPIdir           = fullfile(x.D.PopDir, 'Raw_EPI_Check');
            x.D.T1_ASLREGDIR        = fullfile(x.D.PopDir, 'T1_ASLReg');
            x.D.TTCheckDir          = fullfile(x.D.PopDir, 'ATT_Check');
+			x.D.TExchCheckDir       = fullfile(x.D.PopDir, 'TExch_Check');


Yes - that was the name from Amnah. Am I right @BeatrizPadrela

Functions/Initialization/xASL_init_FileSystem.m

jan-petr · 2022-10-11T21:10:42Z

Modules/SubModule_ASL/xASL_wrp_Quantify.m

@@ -80,6 +80,7 @@ function xASL_wrp_Quantify(x, PWI_Path, pathOutputCBF, M0Path, SliceGradientPath
 iStringCBF = regexpi(pathOutputCBF, 'CBF');
 iStringCBF = iStringCBF(end);
 pathOutputTT = [pathOutputCBF(1:(iStringCBF-1)) 'TT' pathOutputCBF((iStringCBF+3):end)];


I'm fine with that... Up to you. Currently, we just used the current parameter TT.

Functions/Quantification/xASL_quant_Basil.m

Functions/Quantification/xASL_quant_MultiPLD.m

jan-petr · 2022-10-11T21:16:57Z

Modules/SubModule_ASL/xASL_wrp_Quantify.m

    fprintf('%s\n', 'Performing single PLD quantification');
    [~, CBF] = xASL_quant_SinglePLD(PWI, M0_im, SliceGradient, x, x.Q.bUseBasilQuantification); % also runs BASIL, but only in native space!
 	ATT = [];
+	TExch = [];


I've changed that to NaN already in 031559f

jan-petr · 2022-10-11T21:24:06Z

Functions/Quantification/xASL_quant_HadamardDecoding.m

@@ -140,6 +147,19 @@
 	imEncoded = imEncoded(:,:,:,vectorOldOrder);


This is not about Walsh/Hadamard. But about switching order of TEs and PLDs.
So we have original (old order) and reordered (new order).

So we have old/new, or we can have reordered/original etc. I don't think that OLD is particularly bad, but can rename to something else.

HenkMutsaerts

@

Functions/Quantification/xASL_quant_Basil.m

Functions/Quantification/xASL_quant_MultiPLD.m

Modules/SubModule_ASL/xASL_wrp_ResampleASL.m

Modules/xASL_module_ASL.m

HenkMutsaerts

Approved

Modules/SubModule_ASL/xASL_wrp_ResampleASL.m

BeatrizPadrela

Niceee

Functions/Quantification/xASL_quant_Basil.m

Modules/SubModule_ASL/xASL_wrp_ResampleASL.m

Modules/xASL_module_Population.m

Modules/SubModule_Population/xASL_wrp_GetROIstatistics.m

Functions/Initialization/xASL_init_FileSystem.m

BeatrizPadrela · 2022-10-12T07:46:02Z

Functions/Initialization/xASL_init_DefinePaths.m

            x.D.RawEPIdir           = fullfile(x.D.PopDir, 'Raw_EPI_Check');
            x.D.T1_ASLREGDIR        = fullfile(x.D.PopDir, 'T1_ASLReg');
            x.D.TTCheckDir          = fullfile(x.D.PopDir, 'ATT_Check');
+			x.D.TExchCheckDir       = fullfile(x.D.PopDir, 'TExch_Check');


yes she does

… module

…tails

jan-petr requested review from BeatrizPadrela and HenkMutsaerts August 26, 2022 20:13

jan-petr self-assigned this Aug 26, 2022

jan-petr linked an issue Sep 7, 2022 that may be closed by this pull request

FSL/FABBER model for multi-TE quantification #927

Closed

4 tasks

jan-petr commented Sep 7, 2022

View reviewed changes

jan-petr requested review from HenkMutsaerts and removed request for HenkMutsaerts October 5, 2022 18:40

HenkMutsaerts requested changes Oct 11, 2022

View reviewed changes

jan-petr commented Oct 11, 2022

View reviewed changes

jan-petr requested a review from HenkMutsaerts October 12, 2022 14:51

jan-petr force-pushed the feature-#927_FabberQuantificationNew branch from 73abc67 to 203ecc6 Compare October 12, 2022 14:56

HenkMutsaerts requested changes Oct 13, 2022

View reviewed changes

HenkMutsaerts approved these changes Oct 14, 2022

View reviewed changes

Modules/SubModule_ASL/xASL_wrp_ResampleASL.m Show resolved Hide resolved

BeatrizPadrela approved these changes Oct 14, 2022

View reviewed changes

BeatrizPadrela and others added 14 commits October 14, 2022 21:23

#927 Fabber quantification based on parameters from data

13d2428

#927 adjusting the FSL command

cd18290

#927 FSL directory should only change for bMultiTE, not multiPLD

ba1221c

#927 merging quant_Fabber to quant_Basil and adding Basilquant to ASL…

a4eada2

… module

#927 Some adjustements for HAD8 processing

10de559

#927 Hadamard quantification Basil changes

1a37059

#927 Add comments to HadamardDecoding

391a1bc

#927 Inclusion of Texch map in saved files

827c803

#927 TExch in standard space

05c36f4

#927 Fix consistency, headers, and comments

79b020a

#927 Fix path for WSL

5a7d56e

#927 Write PLD option also for single-PLD

ffc260b

#927 xASL_wrp_ResampleASL average repetitions and xASL_quant_Basil de…

90750bf

…tails

#927 Fix PLD and blockSize calculation

279f8f4

jan-petr and others added 13 commits October 14, 2022 21:25

#927 TExch visual QC

991ad25

#927 Fix Hadamard block averaging

9e429fa

#927 correct PLDs for HAD8

9417da4

#927 xASL_quant_Basil Fix PLD averaging

2dd5427

#927 Fix PLD calculation

6c29d7f

#927 Fix missing BASIL output

6fc35de

#927 xASL_wrp_ResampleASL Fix comments and variable names

b41731d

#927 fix PLD averaging

045470a

#927 Use ATT instead of TT for ATT maps

5f89ca0

#927 Allow VisualQC also for multi-PLD

a360cd5

#927 Fix PLD block calculation

a932f29

#927 Rename TExch to Tex

ba69691

#927 xASL_quant_Basil: Rename Texch to Tex

5932913

jan-petr force-pushed the feature-#927_FabberQuantificationNew branch from 4ef10ec to 5932913 Compare October 14, 2022 19:27

jan-petr merged commit 5932913 into develop Oct 14, 2022

jan-petr deleted the feature-#927_FabberQuantificationNew branch October 14, 2022 19:28

		@@ -140,6 +147,19 @@
		imEncoded = imEncoded(:,:,:,vectorOldOrder);

Conversation

jan-petr commented Aug 26, 2022 • edited by HenkMutsaerts Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Linked issue

How to test

Comments

Uh oh!

jan-petr left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jan-petr commented Sep 7, 2022

Uh oh!

jan-petr commented Sep 12, 2022

Uh oh!

HenkMutsaerts left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jan-petr left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

HenkMutsaerts left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

HenkMutsaerts left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

BeatrizPadrela left a comment

jan-petr commented Aug 26, 2022 •

edited by HenkMutsaerts

Loading

HenkMutsaerts left a comment •

edited

Loading

HenkMutsaerts left a comment •

edited

Loading