Skip to content

Queries about running reconstruct_hires_dti.m #6

Description

@dewenmeng

Hello,
I have been using this in order to compute the high resolution DTI from the “low-quality” DTI. The “low-quality” DTI is acquired at the 3T GE scanner in a group of multiple sclerosis patients and healthy controls. I followed the document on GitHub to run the analysis. The compute_model_dti.m is successfully run however there is problems with running reconstruct_hires_dti.m. I used the pretrained tress in the directory trees but I got the error message as below:

Error using NodeRegress (line 20)
Incorrect dimensions for matrix multiplication. Check that the number of columns in the first matrix matches the number of rows in the second matrix. To perform elementwise multiplication, use '.*'.

Error in TreeSuperRes (line 78)
opatchrecon = NodeRegress(tree{tn}, [ipatch_scaled
scale_const]);%tree{tn}.Transformation.M'*ipatch(:);

Error in ForestSuperRes (line 22)
[dtreconrt, accum, accum_var, leafmap] = TreeSuperRes(dt_lowres, trees{i}, n, m, patch_feature_map, us,
'weighted_average_valid', rescale_factor, max_layer);

Error in reconstruct_randomforests (line 68)
parfor dataid = 1:length(data_folders)

Error in reconstruct_hires_dti (line 60)
reconstruct_randomforests(inp_dir, out_dir, train_dir, data_folders, ...

Could you please tell me what the problem is and how can I solve it?

Many thanks,

Dewen

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions