Topology-Preserving Smoothing of Retinotopic Maps

This study aims to directly model the topological condition and generate topology-preserving and smooth retinotopic maps.

• Data is availale here.

---

Structure of project:

| - code/                 // includeing all code supporter the project running.
  | - Libraries/          // there are some third-part libiaries in this folder. 
  | - TPSR/               // this folder is core funtion of our project. 
| - data/                 // data for validation our solutions.
  | - mesh/
  |- ...
| - resources/            // there are some resources in here, like mesh atlas and so on.
| - Figures/              // store some figures from calculation.
| - result/								// store some results after calculation with our approach.

**NOTE: **

• .m files under the root category could run directly and they will show the figure results.

• Data are required when running *figure.m files.

Starting and adding our code to your project.

---

configuration our code to

• Adding the TPSR folder in your project.
• Selecting this folder to the MATLAB path.

start your work

1. data preparation:

• Download our data in above link.
• Add the third part libraries to support some functions, like read_mfile.
• read data file as face, vertices and extra values, there are some samples from our project:

fn = subjects(subi).name;
[Fm,Vm, Em]=read_mfile(['../data/mesh_data/' fn]);

% Load the full hemisphere
[Ffull,Vfull, Efull]=read_mfile(['../data/mesh_data/' fn(1:end-2) '_ecc.m']);

% convert to uvm coordinate with disk_conformal_map()
uvm = disk_conformal_map(Fm,Vm);

% load cut info

roipatch = load('../data/v1');
id2delete = roipatch.id2delete;

[Froi, V_roi, vfather] = gf_remove_mesh_vertices(Fm, Vm, id2delete);

After loading data, you could do some handler continually.

2. Smoothing processing:

Using our functiontopological_smoothingto smooth:

changetol = 0.1; 
smooth_lambda0 = 0.001; % initial of smooth
smooth_avg_k = 2;  % size of smooth along the boundary
meanddth = 1; % average ddth 

visxy_s = topological_smoothing(Froi,uv_p,  visxy_corrected, R2,...
                                anchor, anchorpos,changetol, ...
                            smooth_lambda0,smooth_avg_k, meanddth);

Comparison between approaches of smoothing

results of smoothing V1V2V3

---

• before smoothing

• Average smoothing

• Median smoothing

  

• Laplacian smoothing

• smoothing with our proposal