|
1 | 1 | # Computational Neuroimaging Toolbox |
2 | 2 |
|
3 | | - |
4 | 3 |  |
5 | 4 |  |
6 | 5 |  |
7 | 6 |  |
8 | 7 |
|
9 | | -The Computational Neuroimaging Toolbox is a MATLAB toolbox for estimating input-referred models. Specifically, the toolbox contains tools for Fourier analyses of phase-encoded stimuli, population receptive field mapping, estimating parameters of generic (user-defined) input-referred models as well as performing ridge regression. |
| 8 | +The Computational Neuroimaging Toolbox contains tools for estimating input-referred models. Specifically, it includes tools for Fourier analyses of phase-encoded stimuli, population receptive field (pRF) mapping, fast (real-time) pRF mapping and for estimating parameters of generic (user-defined) input-referred models. |
10 | 9 |
|
11 | 10 | This code is hosted at https://github.com/ccnmaastricht/CNI_toolbox |
12 | 11 | The latest version may always be found here. |
13 | 12 |
|
| 13 | +## Matlab implementation  |
| 14 | + |
14 | 15 | This software was developed with MATLAB R2017a and access to the full suite of MATLAB add-on packages. Some of these packages may be required to run the software. |
15 | 16 |
|
16 | | -## Installation |
17 | | -There are two options for installing the toolbox. Either download the toolbox file [Computational Neuroimaging Toolbox.mltbx](https://github.com/ccnmaastricht/CNI_toolbox/raw/master/Computational%20Neuroimaging%20Toolbox.mltbx), navigate to the downloaded file within MATLAB and then execute the following command: |
| 17 | +### Installation |
| 18 | +Download the toolbox file [Computational Neuroimaging Toolbox.mltbx](https://github.com/ccnmaastricht/CNI_toolbox/raw/master/Computational%20Neuroimaging%20Toolbox.mltbx), navigate to the downloaded file within MATLAB and then execute the following command: |
18 | 19 |
|
19 | 20 | ```MATLAB |
20 | 21 | matlab.addons.toolbox.installToolbox('Computational Neuroimaging Toolbox.mltbx'); |
21 | 22 | ``` |
| 23 | +## Python implementation  |
22 | 24 |
|
23 | | -Alternatively, download the compressed toolbox [Computational Neuroimaging Toolbox.zip](https://github.com/ccnmaastricht/CNI_toolbox/raw/master/Computational%20Neuroimaging%20Toolbox.zip) and extract it into Documents/MATLAB/Add-Ons/Toolboxes. |
24 | | - |
25 | | -## Files |
26 | | -This repository contains four files. |
27 | | -1. `PEA.m` - MATLAB class implementation of Fourier analysis of phase-encoded stimuli. |
28 | | -2. `pRF.m` - MATLAB class implementation of population receptive field mapping. |
29 | | -3. `IRM.m` - MATLAB class implementation of input-referred model estimation. |
30 | | -4. `RRT.m` - MATLAB class implementation of voxel-wise ridge regression. |
31 | | - |
32 | | - |
33 | | -### Phase-encoding analysis tool. |
34 | | -`pea = PEA(parameters)` creates an instance of the PEA class. |
35 | | -parameters is a structure with 7 required fields |
36 | | -- f_sampling: sampling frequency (1/TR) |
37 | | -- f_stim : stimulation frequency |
38 | | -- n_samples : number of samples (volumes) |
39 | | -- n_rows : number of rows (in-plane resolution) |
40 | | -- n_cols : number of columns (in-plance resolution) |
41 | | -- n_slices : number of slices |
| 25 | +### Installation |
42 | 26 |
|
43 | | -This class has the following functions |
44 | | - |
45 | | -- `delay = PEA.get_delay();` |
46 | | -- `direction = PEA.get_direction();` |
47 | | -- `PEA.set_delay(delay);` |
48 | | -- `PEA.set_direction(direction);` |
49 | | -- `results = PEA.fitting(data);` |
50 | | - |
51 | | -Use `help PEA.function` to get more detailed help on any specific |
52 | | -function (e.g. `help PEA.fitting`) |
53 | | - |
54 | | -typical workflow: |
55 | | -```Matlab |
56 | | -pea = PEA(parameters); |
57 | | -pea.set_delay(delay); |
58 | | -pea.set_direction(direction); |
59 | | -results = pea.fitting(data); |
| 27 | +```Python |
| 28 | +pip install cni-tlbx |
60 | 29 | ``` |
61 | | - |
62 | | -### Population receptive field (pRF) mapping tool. |
63 | | -`prf = pRF(parameters)` creates an instance of the pRF class. |
64 | | -parameters is a structure with 7 required fields |
65 | | - - f_sampling: sampling frequency (1/TR) |
66 | | - - n_samples : number of samples (volumes) |
67 | | - - n_rows : number of rows (in-plane resolution) |
68 | | - - n_cols : number of columns (in-plance resolution) |
69 | | - - n_slices : number of slices |
70 | | - - w_stimulus: width of stimulus images in pixels |
71 | | - - h_stimulus: height of stimulus images in pixels |
72 | | - |
73 | | -optional inputs are |
74 | | - - hrf : either a column vector containing a single hemodynamic |
75 | | - response used for every voxel; |
76 | | - or a matrix with a unique hemodynamic response along |
77 | | - its columns for each voxel. |
78 | | - By default the canonical two-gamma hemodynamic response |
79 | | - function is generated internally based on the scan parameters. |
80 | | - |
81 | | -This class has the following functions |
82 | | - |
83 | | - - `hrf = pRF.get_hrf();` |
84 | | - - `stimulus = pRF.get_stimulus();` |
85 | | - - `tc = pRF.get_timecourses();` |
86 | | - - `pRF.set_hrf(hrf);` |
87 | | - - `pRF.set_stimulus(stimulus);` |
88 | | - - `pRF.import_stimulus();` |
89 | | - - `pRF.create_timecourses();` |
90 | | - - `results = pRF.mapping(data);` |
91 | | - |
92 | | -Use `help pRF.function` to get more detailed help on any specific function |
93 | | -(e.g. `help pRF.mapping`) |
94 | | - |
95 | | -typical workflow: |
96 | | -```Matlab |
97 | | -prf = pRF(parameters); |
98 | | -prf.import_stimulus(); |
99 | | -prf.create_timecourses(); |
100 | | -results = prf.mapping(data); |
101 | | -``` |
102 | | - |
103 | | -### Input-referred model (IRM) mapping tool. |
104 | | - |
105 | | -`irm = IRM(parameters)` creates an instance of the IRM class. |
106 | | -parameters is a structure with 5 required fields |
107 | | - - f_sampling: sampling frequency (1/TR) |
108 | | - - n_samples : number of samples (volumes) |
109 | | - - n_rows : number of rows (in-plane resolution) |
110 | | - - n_cols : number of columns (in-plance resolution) |
111 | | - - n_slices : number of slices |
112 | | - |
113 | | -optional inputs are |
114 | | - - hrf : either a column vector containing a single hemodynamic |
115 | | - response used for every voxel; |
116 | | - or a matrix with a unique hemodynamic response along |
117 | | - its columns for each voxel. |
118 | | - By default the canonical two-gamma hemodynamic response |
119 | | - function is generated internally based on the scan parameters. |
120 | | - |
121 | | -This class has the following functions |
122 | | - |
123 | | - - `hrf = IRM.get_hrf();` |
124 | | - - `stimulus = IRM.get_stimulus();` |
125 | | - - `tc = IRM.get_timecourses();` |
126 | | - - `IRM.set_hrf(hrf);` |
127 | | - - `IRM.set_stimulus(stimulus);` |
128 | | - - `IRM.create_timecourses();` |
129 | | - - `results = IRM.mapping(data);` |
130 | | - |
131 | | -Use `help IRM.function` to get more detailed help on any specific function |
132 | | -(e.g. `help IRM.mapping`) |
133 | | - |
134 | | -typical workflow: |
135 | | -```Matlab |
136 | | -irm = IRM(parameters); |
137 | | -irm.set_stimulus(); |
138 | | -irm.create_timecourse(FUN,xdata); |
139 | | -results = irm.mapping(data); |
140 | | -``` |
141 | | - |
142 | | -### Ridge-based analysis tool. |
143 | | - |
144 | | -`rrt = RRT(parameters)` creates an instance of the RRT class. |
145 | | -parameters is a structure with 5 required fields |
146 | | - - f_sampling: sampling frequency (1/TR) |
147 | | - - n_samples : number of samples (volumes) |
148 | | - - n_rows : number of rows (in-plane resolution) |
149 | | - - n_cols : number of columns (in-plance resolution) |
150 | | - - n_slices : number of slices |
151 | | - |
152 | | -optional inputs are |
153 | | - - hrf : either a column vector containing a single hemodynamic |
154 | | - response used for every voxel; |
155 | | - or a matrix with a unique hemodynamic response along |
156 | | - its columns for each voxel. |
157 | | - By default the canonical two-gamma hemodynamic response |
158 | | - function is generated internally based on the scan parameters. |
159 | | - |
160 | | -This class has the following functions |
161 | | - |
162 | | - - `hrf = RRT.get_hrf();` |
163 | | - - `X = RRT.get_design();` |
164 | | - - `RRT.set_hrf(hrf);` |
165 | | - - `RRT.set_design(X);` |
166 | | - - `RRT.optimize_lambda(data,range);` |
167 | | - - `results = RRT.perform_ridge(data);` |
168 | | - |
169 | | -Use `help RRT.function` to get more detailed help on any specific |
170 | | -function (e.g. `help RRT.perform_ridge`) |
171 | | - |
172 | | -typical workflow: |
173 | | -```Matlab |
174 | | -rrt = RRT(parameters); |
175 | | -rrt.set_design(X); |
176 | | -rrt.optimize_lambda(data,range); |
177 | | -results = rrt.perform_ridge(data); |
178 | | -``` |
0 commit comments