OSCR

Enhanced sensorimotor cortex responsiveness to nonplegic hand stimulation and motor network assembly during recovery after spinal cord injury in primates.

Code ↔ Paper

3 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 3 matches
  1. [1] § Methods › Statistical analysis ↔ python/fastECM/fastECM.py, lines 20–146 · score 0.66 · Fast eigenvector centrality, fMRI, node, connected, matrix, mapping
  2. [2] § Methods › Statistical analysis ↔ matlab/fastECM/fastECM.m, lines 1–71 · score 0.64 · Fast eigenvector centrality, fMRI, node, connected, matrix, mapping
  3. [3] § Methods › Statistical analysis ↔ matlab/fastECM/fastECM.m, lines 1–71 · score 0.55 · functional connectivity, Eigenvector centrality, ranks, brain, map

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

MATLAB · 809 lines · 26 KB · GPL-3.0 · 2 matches

  1. function err = fastECM ( inputfile, rankmap, normmap, degmap, maxiter, maskfile, atlasfile, wholemat, dynamics, correlate )
  2. %
  3. % function fastECM ( string <inputfile>,
  4. % bool <rankmap>, bool <normmap>, bool <degmap>,
  5. % int maxiter,
  6. % string <maskfile>, string <atlasfile>,
  7. % bool wholemat,
  8. % int dynamics )
  9. % where - inputfile is the name of a 4D fMRI
  10. % image time series in the NifTI format
  11. % - rankmap ~= 0 produces a file rankECM.nii
  12. % with uniformly distributed [0, 1] centrality (tied) ranks
  13. % - normmap ~= 0 produces a file normECM.nii
  14. % with normally distributed N (0, 1) centralities generated from ranks
  15. % - degmap ~= 0 produces a node power map (similar to degree)
  16. % this is the column sum of connection strengths
  17. % - maxiter limits the number of iterations of the algorithm
  18. % - maskfile selects voxels inside a mask
  19. % - atlasfile groups signals by pre-defined regions
  20. % - wholemat: do not use the 'fast' recipe, write out
  21. % connectivity matrix and compute graph measures (costly!)
  22. % - dynamics: number of ECM to extract from a 4D fMRI
  23. % - correlate: way to approximate the eigenvector of
  24. % correlated time series:
  25. % 'fast' -- default, correlation + 1
  26. % 'relu' -- ReLU, rectified linear unit
  27. %
  28. % returns a file fastECM.nii in the same directory as <inputfile>
  29. % which contains a 'fast eigenvector centrality mapping'
  30. % whole-brain voxelwise functional connectivity analysis;
  31. % error code 0 if all is well, 1 otherwise
  32. %
  33. %
  34. %
  35. % Example 1
  36. %
  37. % If the following files are present:
  38. % fmri4d.nii.gz -- containing an fMRI time series
  39. % mask_csf.nii.gz -- mask with non-brain tissue and CSF removed
  40. % aal_MNI_V4_4mm_gong.nii.gz -- containing a volume with atlas labels
  41. % the call
  42. % >> fastECM;
  43. % runs a demo analysis highlighting all these options
  44. %
  45. % Example 2
  46. %
  47. % For using some, but not all options, the easiest way to
  48. % call fastECM is to pass these options in a struct:
  49. % >> op.inputfile = 'fmri4d.nii.gz';
  50. % >> op.dynamics = 25;
  51. % >> fastECM (op);
  52. %
  53. %
  54. %
  55. % (c) Alle Meije Wink -- 16/03/2012
  56. % a.m.wink AT gmail.com
  57. %
  58. %
  59. %
  60. % If you use this method in your research, remember to cite this paper
  61. % in the journal "Brain Connectivity":
  62. %
  63. % Alle Meije Wink, Jan C de Munck, Ysbrand D van der Werf, Odile A van den heuvel, Frederik Barkhof
  64. % "Fast eigenvector centrality mapping of voxel-wise connectivity in functional MRI:
  65. % implementation, validation and interpretation."
  66. % Brain Connectivity 2012, Vol. 2 No. 5, pages 265-274
  67. % URL: http://online.liebertpub.com/doi/abs/10.1089/brain.2012.0087
  68. % or: http://dare.ubvu.vu.nl/handle/1871/48750
  69. %
  70. %
  71. %
  75. % If called w/o agruments, produce a demo call
  76. if (~nargin)
  78. inputfile = [fileparts(which('fastECM.m')) filesep 'fmri4d.nii.gz' ];
  79. maskfile = [fileparts(which('fastECM.m')) filesep 'mask_csf.nii.gz' ];
  80. atlasfile = [fileparts(which('fastECM.m')) filesep 'aal_MNI_V4_4mm_gong.nii.gz' ];
  82. fprintf (2, ...
  83. ['\n syntax:\n fastECM (\n \t<str inputfile>, \n \t<bool rankmap>, <bool normmap>, ' ...
  84. '<bool degmap>, \n \t<int maxiter>, \n \t<str maskfile>, <str atlasfile>,' ...
  85. '\n \t<bool wholemat>,\n \t<int dynamics>,\n \t<str correlate>\n \t\b) \n\n']);
  87. correlate = 'fast';
  89. fprintf (' inputfile = ''%s'';\n', inputfile);
  90. fprintf (' maskfile = ''%s'';\n', maskfile);
  91. fprintf (' atlasfile = ''%s'';\n', atlasfile);
  92. fprintf (' correlate = ''%s'';\n', correlate);
  94. democall = ['fastECM ( inputfile, 1, 1, 1, 20, maskfile, atlasfile, 0, 25, correlate );'];
  95. fprintf ('\n %% demo call:\n >> %s\n\n', democall);
  97. if ( (exist (inputfile) *exist (maskfile) *exist (atlasfile) ) ~= 8)
  99. inputfile = inputfile (1: (end-3) );
  100. maskfile = maskfile (1: (end-3) );
  101. atlasfile = atlasfile (1: (end-3) );
  103. if (exist (inputfile) ~= 2)
  105. fprintf ('warning: not all demo files found, exiting\n');
  107. err = 1;
  108. return;
  110. end % if ~input files (uncompressed)
  112. end % if ~inputfile or ~maskfile or ~atlasfiles
  114. err = eval (democall);
  115. fprintf ('\n');
  117. return; % call fastECM inside this call
  119. else
  121. err = 0; % continue without errors
  123. end % if ~nargin
  127. % check for options for no. of dynamics,
  128. % whole matrix computation,
  129. % atlas region selection,
  130. % mask selection
  131. % max. iterations
  132. % write degree (node power) map
  133. % write normally distributed map
  134. % write ranks (uniformly distributed) map
  135. % and set defaults
  136. if (nargin<10)
  137. correlate = 'fast';
  138. if (nargin<9)
  139. dynamics = 1;
  140. if (nargin<8)
  141. wholemat = 0;
  142. if (nargin<7)
  143. atlasfile = 0;
  144. if (nargin<6)
  145. maskfile = 0;
  146. if (nargin<5)
  147. maxiter = 25;
  148. if (nargin<4)
  149. degmap = 0;
  150. if (nargin<3)
  151. normmap = 0;
  152. if (nargin<2)
  153. rankmap = 0;
  154. end % if rankmap
  155. end % if normmap
  156. end % if degmap
  157. end % if maxiter
  158. end % if maskfile
  159. end % if atlasfile
  160. end % if wholemat
  161. end % if dynamics
  162. end % if correlate
  166. % if options were given as a struct
  167. % use the existing fields
  168. if ( isstruct (inputfile) )
  170. input_parameters = inputfile;
  171. fieldnames = { 'rankmap' 'normmap' 'degmap' 'maxiter' 'maskfile' 'atlasfile' 'wholemat' 'dynamics' 'correlate' 'inputfile' };
  173. for f = 1:length (fieldnames)
  175. if isfield ( input_parameters, fieldnames{f} )
  176. eval ([ fieldnames{f} ' = input_parameters.' fieldnames{f} ';' ]);
  177. end % if isfield
  179. end % for f
  181. if isstruct (inputfile)
  182. fprintf ('error: ''inputfile'' not defined in struct, exiting\n');
  183. err = 1;
  184. return;
  185. end % if isstruct
  187. end % if isstruct
  191. % check if nifti support exists
  192. % otherwise add fastECM-supplied version
  193. if (exist ('load_untouch_nii') ~= 2)
  195. fprintf (2,'Software for reading NifTI images not found. \n');
  196. fprintf (2,'Using Tools for Nifti/Analyze for NifTI file I / O. \n');
  197. fprintf (2,'www.mathworks.com/matlabcentral/fileexchange/8797\n\n');
  198. usenifti = 1;
  199. npath = [fileparts(which('fastECM.m')) filesep 'tools4nifti'];
  200. addpath (npath);
  202. else
  204. usenifti = 0;
  206. end % if exist
  210. % load NifTI input file
  211. fprintf ('reading %s ...\n', inputfile);
  213. if (inputfile (1) ~= filesep) % if path does not start with a separator
  215. if ( (isunix) | ... % and -in windows- not with a drive letter or '\\'
  216. ( (~isunix) & (inputfile (2) ~= '\') & (inputfile (2) ~= ':') ) ...
  217. ) % make it an absolute path
  219. inputfile = [pwd filesep inputfile];
  221. end % if isunix
  223. end % if inputfile
  225. % get the directory, base name and extension (s)
  226. [fd, fn, fx2] = fileparts (inputfile); % dir and base file name + extension (may be combined)
  227. [fd1, fn, fx] = fileparts (fn); % base file name and 1st extension (2nd may be .gz)
  229. % get the data and clear data from file record
  230. M = load_untouch_nii (inputfile); % read the information from the NifTI file
  231. m = double (M.img); % get the 4d voxel data
  232. M.img = []; % empty img after reading
  233. M = rmfield (M, 'img'); % and remove from M
  234. msz = size (m);
  235. tln = msz (4); % store the time series length
  236. msz = msz (1:3); % store the size of one image volume
  237. m (find (~isfinite (m) ) ) = 0; % get rid of NaNs
  239. % check whether a mask filename has been given and if dimensions are OK
  240. msk = 1; % default value for no user-supplied mask
  241. if ( maskfile ~= 0 & ~isempty (maskfile) )
  243. if (maskfile (1) ~= filesep) % if path does not start with a separator
  245. if ( (isunix) | ... % and -in windows- not with a drive letter or '\\'
  246. ( (~isunix) & (maskfile (2) ~= '\') & (maskfile (2) ~= ':') ) ...
  247. ) % make it an absolute path
  248. maskfile = [pwd filesep maskfile];
  249. end % if isunix
  251. end % if maskfile
  253. Msk = load_untouch_nii (maskfile); % read the information from the NifTI file
  254. msk = double (Msk.img); % get the 3d voxel data
  255. Msk.img = []; % empty img after reading
  257. if (size (msk) ~= msz) % test if mask volume is the right size
  259. warning (sprintf (['dimensions of %s incompatible with %s\n' ...
  260. 'continuing with nonzero time series.'], ...
  261. maskfile, inputfile ) );
  262. mskfile = 0;
  263. msk = 1;
  265. end % if size
  267. else
  269. mskfile = 0;
  271. end % if maskfile
  273. % if the mask provided is not OK then the positions of nonzeros in the 1st volume
  274. msk = msk.*squeeze (abs (min (m, [], 4) ) ); % take the minimum of each timeseries
  275. msk = (msk>0); % make mask binary
  277. % if atlas is used, generate regional time series
  278. atl = 1; % default value for no user-supplied atlas
  279. if ( atlasfile ~= 0 & ~isempty (atlasfile) )
  281. if (atlasfile (1) ~= filesep) % if path does not start with a separator
  283. if ( (isunix) | ... % and in windos not with a drive letter or '\\'
  284. ( (~isunix) & (atlasfile (2) ~= '\') & (atlasfile (2) ~= ':') ) ...
  285. ) % make it an absolute path
  286. atlasfile = [pwd filesep atlasfile];
  287. end % if isunix
  289. end % if atlasfile
  291. Atl = load_untouch_nii (atlasfile); % read the information from the NifTI file
  292. atl = double (Atl.img); % get the 3d voxel data of atlas regions
  293. Atl.img = []; % empty img after reading
  295. if (size (atl) ~= msz) % test if atlas volume is the right size
  297. warning (sprintf (['dimensions of %s incompatible with %s\n' ...
  298. 'continuing by using all voxel time series'], ...
  299. atlasfile, inputfile ) );
  300. atlasfile = 0;
  301. atl = 1;
  303. else
  305. msk = msk.* (atl>0);
  307. end % if size
  309. else
  311. atlasfile = 0;
  313. end % if atlasfile
  315. fprintf ('\nloaded and masked\n');
  319. % make time series 2D [tln msk]
  320. m = reshape(m, [prod(msz) tln])' ; % reshape the 4d voxels as 2d: [tln msz]
  321. m = double(m (:, find (msk) ) ); % continue with only the nonzero voxels
  325. % if atlasfile found -> make regional time series
  326. if ( ~atlasfile )
  328. np = size (m, 2); % store the number of nonzero voxels
  330. else
  332. fprintf ('atlasing\n');
  334. mreg = max (atl (:) ); % highest region label
  335. atl = atl (find (msk) ); % atl now the same size as msk
  336. reg = unique (atl); % region labels: nonzero values found inside mask/atl
  338. mm = zeros (tln, length (reg) );
  339. for r = reg (:)' % construct regional means
  340. mm (:, r) = mean ( m (:, find (atl == r) ), 2 );
  341. end % for r
  342. m = mm; % continue with regional mean time series
  343. clear mm;
  345. txtfile = [fd filesep fn '_fastECMtseries.txt'];
  346. dlmwrite (txtfile, m, ' '); % write regional mean time series to text file
  348. matfile = [fd filesep fn '_fastECMtseries.mat'];
  349. save (matfile, 'm');
  351. np = r; % store the number of included regios
  353. wholemat = 1; % regional matrices are small -> provide all info
  355. end % if ~atlasfile
  359. % compute mean and var
  360. mav = mean (m); % compute the time series mean
  361. mvr = std (m); % compute the time series standard deviation
  362. mvr = mvr+eps; % prevent divisions by 0
  366. % prepare matrix
  367. m = (m - (ones (tln, 1) *mav) ) ...
  368. ./ (ones (tln, 1) *mvr); % mean 0, std 1 to make covariance matrix
  369. m = m/sqrt (tln-1); % make correlations instead (diagonal 1)
  373. % compute ECM on intervals given by the number of dynamics
  374. ddiff = tln-dynamics;
  375. for d = 1:dynamics
  377. % initialise eigenvector estimate v
  378. vprev = 0; % 'initialise' previous ECM estimate
  379. vcurr = ones (np, 1) /sqrt (np); % initialise estimate with L2-norm == 1
  381. iter = 0; % reset iteration counter
  382. dnorm = 1; % initial value for difference L2-norm
  383. cnorm = 0; % initial value for estimate L2-norm
  385. m0 = m (d: (ddiff+d), :); % use the interval for the current dynamic
  387. if (strcmp (correlate, 'relu') ) % use ReLU equation to guarantee positive M
  389. m0 = [m0; abs(m0) ]; % concatenate absolute value of time series
  391. end % if strcmp
  393. % efficient power iteration for correlated time series
  394. while ( (iter<maxiter) & (dnorm>cnorm) )
  396. vprev = vcurr; % start with previous estimate
  397. prevsum = sum (vprev); % sum of estimate
  399. if (~wholemat) % A. 'fast recipe' -> cunning re-ordering of computations
  401. vcurr_1 = m0*vprev; % 1. part one of M*v
  402. vcurr_2 = m0'*vcurr_1; % 2. part two of M*v
  404. % if ReLU correlation is used, then values are positive (modulo
  405. % some precision errors). Otherwise, this needs to be enforced
  406. if (strcmp (correlate, 'relu') )
  408. vcurr_3 = vcurr_2 .* (vcurr_2>0); % 3. remove remaining small negative values
  410. else
  412. vcurr_3 = vcurr_2+prevsum; % 3. adding sum -- same effect as [M+1]*v
  414. end % if strcmp
  416. else % B. original recipe -> provide full connectivity info
  418. vcurr_1 = (m0'*m0); % 1. M = correlations
  419. vcurr_2 = vcurr_1*vprev; % 2. M * v
  421. % if ReLU correlation is used, then values are positive (modulo
  422. % some precision errors). Otherwise, this needs to be enforced
  423. if (strcmp (correlate, 'relu') )
  425. vcurr_3 = vcurr_2+1; % 3. remove remaining small negative values
  427. else
  429. vcurr_3 = vcurr_2+1; % 3. [M+1]*v
  431. end
  433. if (~iter) % compute & write graph measures only in iteration 1
  435. vcurr_1 = vcurr_1-diag (diag (vcurr_1) );
  436. connmat_out (:, :, d) = vcurr_1;
  437. dist = 1./vcurr_1; % distance matrix <-> 1/connectivity
  439. for c = 1:np % shortest distances www.ee.columbia.edu/~marios/matlab/tips.pdf
  440. dist = min (dist, repmat (dist (:, c), [1 np]) +repmat (dist (c, :), [np 1]) );
  441. end % for c
  443. mpl = mean (dist (dist ~= inf) ); % average shortest path to other nodes -> global path length
  445. % threshold & binarise for other graph measures
  446. lv = length (vcurr_1);
  447. indi = triu (reshape (1: (lv*lv), [lv lv]), 1);
  449. if (exist ('backbone_wu') ~= 2); % get 'backbone' i.e. MST + some extra
  451. eval (['!wget -P ' fileparts(which ('fastECM.m')) ...
  452. ' -O backbone_wu.m ' ...
  453. 'sites.google.com/site/bctnet/Home/functions/backbone_wu.m']);
  455. end % if exist
  457. no0corr = vcurr_1; % make correlation matrix where 0-pairs have very low, non-zero correlation
  458. no0corr (~no0corr) = min (no0corr (:) ) -eps; % otherwise the BCT script will not add these nodes to the MST
  459. no0corr = no0corr+1;
  460. no0corr (find (diag (diag (no0corr) ) ) ) = 0;
  462. % compute MST and add nodes up to degree sqrt (#nodes) to the 'backbone'
  463. [mst clus] = backbone_wu (no0corr, fix (sqrt (lv) ) );
  464. vcurr_bin (:, :, d) = sign (clus);
  465. vcurr_mst (:, :, d) = sign (mst);
  467. % apply bct measures to vcurr_bin: communities (index per node)
  468. if (exist ('community_louvain') ~= 2);
  470. eval (['!wget -P ' fileparts(which ('fastECM.m') ) ...
  471. ' -O community_louvain.m ' ...
  472. 'sites.google.com/site/bctnet/Home/functions/community_louvain.m ']);
  474. end % if exist
  475. communities (:, d) = community_louvain (vcurr_bin (:, :, d) );
  477. % apply bct measures to vcurr_bin: betweenness (per node)
  478. if (exist ('betweenness_bin') ~= 2);
  480. eval (['!wget -P ' fileparts(which ('fastECM.m') ) ...
  481. ' -O betweenness_bin.m ' ...
  482. 'sites.google.com/site/bctnet/Home/functions/betweenness_bin.m ']);
  484. end % if exist
  485. betweenness (:, d) = betweenness_bin (vcurr_bin (:, :, d) );
  487. % apply bct measures to vcurr_bin: clustering (per node)
  488. if (exist ('clustering_coef_bu') ~= 2);
  490. eval (['!wget -P ' fileparts(which('fastECM.m') ) ...
  491. ' -O clustering_coef_bu.m ' ...
  492. ' sites.google.com/site/bctnet/Home/functions/clustering_coef_bu.m ']);
  494. end % if exist
  495. clustering (:, d) = clustering_coef_bu (vcurr_bin (:, :, d) );
  497. % apply bct measures to vcurr_bin: path length (per node)
  498. if (exist ('distance_bin') ~= 2);
  500. eval (['!wget -P ' fileparts(which ('fastECM.m') ) ...
  501. ' -O distance_bin.m ' ...
  502. ' sites.google.com/site/bctnet/Home/functions/distance_bin.m ']);
  504. end % if exist
  505. pathlengthtmp = distance_bin (vcurr_bin (:, :, d) );
  506. pathlength (:, d) = mean (pathlengthtmp);
  508. end % if ~iter
  510. end % if wholemat
  512. vcurr = vcurr_3/norm (vcurr_3, 2); % normalise L2-norm
  514. if ( (~iter) & (degmap ~= 0) )
  516. dvcurr (:, d) = vcurr_2 (:); % save 'node power' of this dynamic if requested
  517. end % if degmap
  519. iter = iter+1; % increase iteration counter
  520. dnorm = norm (vcurr-vprev, 2); % L2-norm of difference prev-curr estimate
  521. cnorm = norm (vcurr, 2) *eps; % L2-norm of current estimate
  522. fprintf ('dynamic %04d (%04d - %04d), iteration %04d, || v_i - v_ (i-1) || / || v_i * epsilon || = %0.16f / %0.16f\r', ...
  523. d, d-1, ddiff+d-1, iter, dnorm, cnorm)
  525. end % while
  527. if ( (rankmap ~= 0) | (normmap~= 0) )
  529. rvcurr (:, d) = tiedrank (vcurr) ... % tied ranks: equal values lead to equal ranks
  530. / (length (vcurr) +1); % division: from uniform [1, N] to uniform ]0, 1[
  532. if ( normmap ~= 0)
  534. mu = 0; % mean for N (0, 1) distributed centralities
  535. sig = 1; % standard deviation for N (0, 1) distribution
  537. % produce a map of gaussianised EC ranks if requested and write to normECM
  538. nvcurr (:, d) = mu+sqrt (2) *sig ... % probit function (en.wikipedia.org/wiki/Probit)
  539. * erfinv ... % based on inverse error function
  540. (2*rvcurr (:, d) -1); % uniform ]0, 1[ to N (0, 1) via inverse transform sampling
  542. end %if (normmap)
  544. end % if (rankmap)
  546. vcurr_out (:, d) = vcurr;
  547. fprintf ('\n');
  549. end % for d
  553. % write the eigenvactor centrality map to the file fastECM.nii (.gz)
  554. % and write orther centralities (node power, uniform, normal) if requested
  555. write_map (inputfile, M, msk, atl, vcurr_out, 'fastECM', sprintf ('ECM [iterations: %d]', iter) );
  556. if ( exist ('dvcurr') == 1 )
  558. write_map (inputfile, M, msk, atl, dvcurr, 'degCM', 'weighted degree centrality (node power) ');
  560. end % if dvcurr
  562. if ( (exist ('rvcurr') == 1) & rankmap )
  564. write_map (inputfile, M, msk, atl, rvcurr, 'rankECM', 'ECM [converted to ]0, 1[ ranks]');
  566. end % if rvcurr
  568. if ( exist ('nvcurr') == 1 )
  570. write_map (inputfile, M, msk, atl, nvcurr, 'normECM', 'ECM [converted to N (0, 1) values]');
  572. end % if nvcurr
  576. % if graph measures have been computed -> write them to a spreadsheet-readable
  577. % XML-file (take care that native XML may take a long time to load even for small files)
  578. if (exist ('connmat_out') == 1)
  580. % put matrices also in [vector_per_volume #volumes] format
  581. connmat_out = reshape(connmat_out, [prod(size(clus)) dynamics]);
  582. vcurr_bin = reshape(vcurr_bin, [prod(size(clus)) dynamics]);
  583. vcurr_mst = reshape(vcurr_mst, [prod(size(clus)) dynamics]);
  585. write_map (inputfile, M, ones (size (clus) ), 1, connmat_out, 'connections', 'connectivity matrix');
  586. write_map (inputfile, M, ones (size (clus) ), 1, vcurr_bin, 'backbone', 'binary backbone');
  587. write_map (inputfile, M, ones (size (clus) ), 1, vcurr_mst, 'min_span', 'minimal spanning tree');
  588. write_map (inputfile, M, msk, atl, communities, 'communities', 'community_louvain');
  589. write_map (inputfile, M, msk, atl, betweenness, 'betweenness', 'betweenness');
  590. write_map (inputfile, M, msk, atl, clustering, 'clustering', 'clustering');
  591. write_map (inputfile, M, msk, atl, pathlength, 'path_length', 'pathlength');
  593. % also make readable files: .mat and office-compatible XML
  594. matfile =[fd filesep fn '_fastECMstats.mat'];
  595. xmlfile = [fd filesep fn '_fastECMstats.xml'];
  596. fastECMstats.fastECM = vcurr_out;
  597. fid = fopen (xmlfile, 'w');
  598. if (fid> (-1) )
  600. fprintf (fid, '<?xml version = "1.0" encoding = "UTF-8" standalone = "yes"?>\n');
  601. fprintf (fid, ['<ss:Workbook xmlns:ss = "urn:schemas-microsoft-com:office:spreadsheet">\n']);
  602. writetosheet (fid, vcurr_out, 'fastECM');
  604. if ( exist ('dvcurr') == 1 )
  606. writetosheet (fid, dvcurr, 'degCM');
  607. fastECMstats.degCM = dvcurr;
  609. end
  611. if ( (exist ('rvcurr') == 1) & rankmap )
  613. writetosheet (fid, rvcurr, 'rankECM');
  614. fastECMstats.rankECM = rvcurr;
  616. end
  618. if ( exist ('nvcurr') == 1 )
  620. writetosheet (fid, nvcurr, 'normECM');
  621. fastECMstats.normECM = nvcurr;
  623. end
  625. % these variables cannot be written to spreadsheet
  626. % for some reason, but they can be saved as a .mat
  627. %
  628. %writetosheet (fid, connmat_out, 'connections'); % crashes spreadsheet program
  629. fastECMstats.connections = connmat_out;
  630. %writetosheet (fid, vcurr_bin, 'backbone'); % crashes spreadsheet program
  631. fastECMstats.backbone = vcurr_bin;
  633. writetosheet (fid, communities, 'communities');
  634. fastECMstats.communities = communities;
  635. writetosheet (fid, betweenness, 'betweenness');
  636. fastECMstats.betweenness = betweenness;
  637. writetosheet (fid, clustering, 'clustering');
  638. fastECMstats.clustering = clustering;
  639. writetosheet (fid, pathlength, 'path_length');
  640. fastECMstats.path_length = pathlength;
  641. fprintf (fid, '</ss:Workbook>\n');
  642. fclose (fid);
  644. % write fastECMstats to a .mat file
  645. save (matfile,'fastECMstats');
  647. end % if fid
  649. end % if connmat_out
  651. % if nifti support was added before, remove it again to not leave prints
  652. if (usenifti)
  654. rmpath (npath); % was own nifti added to path -- if yes then remove now
  656. end % if usenifti
  658. return % fastECM
  660. % tiedrank - Tied rank of each element in a set
  661. % [r] = tiedrank (X, dim)
  662. % Computes tied rank of each element of X along given dimension
  663. % (default is to use the first non-singleton dimension)
  664. % "Tied ranking" is such that ex-aequo elements share the same (possibly half) rank:
  665. % >> tiedrank (['DABBC']) %-> 6 1 2.5 2.5 4
  666. % from https://github.com/kndiaye/matlab/blob/master/tiedrank.m
  667. function [r] = tiedrank (X, dim)
  669. if nargin<2
  671. dim = min (find (size (X) >1) );
  673. if isempty (dim)
  675. error ('X is empty!')
  677. end % if isempty
  679. end % if nargin
  681. [ignore, Y] = sort (X, dim);
  682. [ignore, r1] = sort (Y, dim);
  683. [ignore, Y] = sort (-X, dim);
  684. [ignore, r2] = sort (Y, dim);
  685. r2 = size (X, dim) -r2+1;
  686. r = (r1+r2) /2;
  688. return
  690. % function to write a variable as an XML sheet
  691. % the option of writing flattened upper triangular matrices is implemented but never used!
  692. function writetosheet (fid, variable, sheetname);
  694. fprintf (fid, '<ss:Worksheet ss:Name = "%s">\n<ss:Table>\n', sheetname);
  696. if (length (size (variable) ) == 3)
  698. for m = 1:size (variable, 3) % write matrix by flattening upper triangular matrix
  700. variable2d (:, m) = nonzeros (triu (squeeze (variable (:, :, m) ), 1) );
  702. end % for m
  703. variable = variable2d;
  704. clear variable2d;
  706. end % if length
  708. for c = 1:size (variable, 2)
  710. fprintf (fid, '<ss:Row>\n');
  712. for r = 1:size (variable, 1)
  714. fprintf (fid, '<ss:Cell><ss:Data ss:Type = "Number">%0.8f</ss:Data></ss:Cell>', variable (r, c) );
  716. end % for r
  718. fprintf (fid, '</ss:Row>\n');
  720. end % for c
  722. fprintf (fid, '</ss:Table>\n</ss:Worksheet>\n');
  724. return
  728. % function to write a map/matrix as a nifti file
  729. function write_map (inputfile, M, msk, atl, vcurr, mapfile, descrip)
  730. %
  731. % writes a map based on a template file
  732. % - inputfile = filename of the template
  733. % - M = nifti record of the template
  734. % - msk = brain mask of the template
  735. % - atl = atlas volume (if used, otherwise 1)
  736. % - vcurr = map values inside brain mask
  737. % - mapfile = filename of the map file
  738. % - descrip = description of contents
  739. %
  740. %
  741. %
  742. % this routine uses the following external code
  743. %
  744. % " Tools for Nifti / Analyze " by Jimmy Shen
  745. % http://www.mathworks.com/matlabcentral/fileexchange/8797
  746. % " quantile.m " by Anders Holtsberg
  747. % www.spatial-econometrics.com/distrib/quantile.m
  748. %
  752. % if atlas used, make a map of regional centralities
  753. if (prod (size (atl) ) ~= 1)
  755. vcurr_2 = zeros (size (atl) );
  756. reg = unique (atl); % region labels: nonzero values found inside atl
  757. for d = 1:size (vcurr, 2)
  759. for r = 1:length (reg)
  761. nvox = find (atl == reg (r) );
  762. vcurr_2 (nvox, d) = vcurr (r, d);%/ (length (nvox) );
  764. end % for r
  766. end % for d
  768. vcurr = vcurr_2;
  769. clear vcurr_2;
  771. end % if prod
  773. % create output array mout (and initialise 0)
  774. mout = zeros ([size(msk) size(vcurr, 2)]);
  775. mout = nan*mout; % NaN outside mask
  777. qnts = quantiles (vcurr ...
  778. (vcurr>0), ...
  779. [.01 .99]); % quantiles to set contrast
  781. % put vcurr in mout
  782. smout = size (mout);
  783. mout = reshape (mout, [prod(size(msk)) size(vcurr, 2)]);
  784. mout (find (msk), :) = vcurr;
  785. mout = reshape (mout, smout);
  787. % write output nifti file based on mout
  788. [fd, fn, fx2] = fileparts (inputfile);
  789. [fd1, fn, fx] = fileparts (fn); % base file name and 1st extension (2nd may be .gz)
  790. fx = [fx fx2]; % in which case, concatenate by adding fx2
  791. outputfile = [fd filesep fn '_' mapfile fx];
  792. fprintf ('writing %s ...\n', outputfile);
  793. M.hdr.hist.descrip = sprintf ('generated by fastECM - %s', descrip);
  794. M.hdr.dime.dim = [ length(size(mout)) size(mout) ];
  795. M.hdr.dime.dim ( (end+1) :8) = 1; % map is not 4D
  796. M.hdr.dime.pixdim ( (end+1) :8) = 1; % neither are its voxels
  797. if ~max ([32, 64] == M.hdr.dime.bitpix) % if data type not at least float
  798. M.hdr.dime.datatype = 64; % make it double
  799. end;
  800. M.hdr.dime.cal_min = qnts (1); % min of range (for win/lev)
  801. M.hdr.dime.cal_max = qnts (2); % max
  802. M.hdr.dime.glmin = qnts (1); % min of range
  803. M.hdr.dime.glmax = qnts (2); % max
  804. M.hdr.dime.scl_slope = 1; % the quantiles only make sense
  805. M.hdr.dime.scl_inter = 0; % with slope 1 and intercept 0
  806. M.img = mout; % add voxel data to map
  807. save_untouch_nii (M, outputfile); % write the file
  809. return

fastECM.m at commit d0c2b7e, under GPL-3.0 · at the source

Overview

Authors: Takamichi Tohyama1,2, Reona Yamaguchi3,4, Naokazu Goda1,5,6, Tetsuya Yamamoto1,5,7, Norihiro Sadato5,7,8, Tadashi Isa3,4,5,9, Masaki Fukunaga1,5,6
  1. Section of Brain Function Information, National Institute for Physiological Sciences,38 Nishigonaka, Myodaiji, Okazaki, 444-8585 Japan
  2. Department of Rehabilitation Medicine, School of Medicine, Fujita Health University,1-98 Dengakugakubo, Kutsukake, Toyoake, 470-1192 Japan
  3. Institute for the Advanced Study of Human Biology (WPI-ASHBi), Kyoto University,Yoshida-konoe-cho, Sakyo-ku, Kyoto, 606-8501 Japan
  4. Department of Neuroscience, Graduate School of Medicine, Kyoto University,Yoshida-konoe-cho, Sakyo-ku, Kyoto, 606-8501 Japan
  5. Core for Spin Life Sciences, Okazaki Collaborative Platform, National Institutes of Natural Sciences,38 Nishigonaka, Myodaiji, Okazaki, 444-8585 Japan
  6. Physiological Science Program, Graduate Institute for Advanced Studies, SOKENDAI, Shonan Village,Hayama, 240-0193 Japan
  7. Research Organization of Science and Technology, Ritsumeikan University,1-1-1 Noji-Higashi, Kusatsu, 525-8577 Japan
  8. Center for Research Collaboration, National Institute for Physiological Sciences,38 Nishigonaka, Myodaiji, Okazaki, 444-8585 Japan
  9. Section of Advanced Project Promotion, National Institute for Physiological Sciences,38 Nishigonaka, Myodaiji, Okazaki, 444-8585 Japan
Institutions: National Institute for Physiological Sciences (Japan); Fujita Health University (Japan); Kyoto University (Japan); National Institutes of Natural Sciences (Japan); The Graduate University for Advanced Studies, SOKENDAI (Japan); Ritsumeikan University (Japan)
Journal: Scientific reports, volume 16, issue 1, article 24253
Dates: received 27 March 2025; accepted 19 May 2026; published online 27 May 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1038/s41598-026-54528-7 · PMID 42204263 · PMCID PMC13443211 · OpenAlex W7162523560
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: fMRI (modality), non-human primate (organism), other condition (population), systems (subfield)
Methods: Spectral & time-frequency, Statistics, Preprocessing, Connectivity, fMRI & imaging, Smoothing, state filtering, decompositions, Physiology & signal measures
Keywords: Neuronal network, Brain plasticity, Functional neuroimaging, Sensorimotor area, Motor cortex, Premotor cortex, Cortex, Spinal cord injury, Functional magnetic resonance imaging
MeSH: Hand*, Motor Cortex*, Recovery of Function*, Sensorimotor Cortex*, Spinal Cord Injuries*, Animals, Magnetic Resonance Imaging, Male (* major topic)
Topic: Transcranial Magnetic Stimulation Studies (Neurology, Neuroscience), according to OpenAlex
Funding: Japan Society for the Promotion of Science (JP22K15624, JP22H04992, JP19K22985); Japan Agency for Medical Research and Development (JP18dm0307005, JP17dm0107118)
Citations: not cited yet (Europe PMC); 55 references in the paper

Abstract

The abstract is not reproduced here: the paper's license (CC BY-NC-ND) does not allow it. Read it in the paper, at the publisher or on Europe PMC.

Repository

Its files are read in the Code ↔ Paper reader above, with 3 matches between paragraphs and lines of code.

amwink/bias

License: GPL-3.0
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: d0c2b7ecd16b0c6f638a00c596c59b899fc13e0b, 25 August 2022
Languages: MATLAB (132), C (13), Shell (6), Python (5), C/C++ (2), C++ (1)
Size: 186 files, 159 scripts
Software Heritage: archived
Found in: the text, “Statistical analysis”
Holds: README, license file
Not found: CITATION.cff, environment file, tests, continuous integration, documentation
Tools: Tools for NIfTI and ANALYZE image (MATLAB) (15 files), SPM (14 files), FSL (4 files), NiBabel (3 files), NumPy (3 files), Brain Connectivity Toolbox (1 file), DIPY (1 file), Image Processing Toolbox (1 file), Matplotlib (1 file), pydicom (1 file), SciPy (1 file)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
161 files
At the source: github.com/amwink/bias/tree/master/matlab/fastECM

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 159 scripts, each with its path and the digest of its content;
  • 3 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data availability statement

The paper has a data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

  • it says that the data are available on request

Read it in the paper: doi.org/10.1038/s41598-026-54528-7.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 7 authors, 9 keywords, 8 MeSH terms, 2 funders, 55 references.

Cite

This paper

Tohyama, T., Yamaguchi, R., Goda, N., Yamamoto, T., Sadato, N., Isa, T., & Fukunaga, M. (2026). Enhanced sensorimotor cortex responsiveness to nonplegic hand stimulation and motor network assembly during recovery after spinal cord injury in primates. Scientific reports, 16(1), 24253. https://doi.org/10.1038/s41598-026-54528-7

BibTeX

@article{tohyama2026enhanced,
author = {Tohyama, Takamichi and Yamaguchi, Reona and Goda, Naokazu and Yamamoto, Tetsuya and Sadato, Norihiro and Isa, Tadashi and Fukunaga, Masaki},
title = {{Enhanced sensorimotor cortex responsiveness to nonplegic hand stimulation and motor network assembly during recovery after spinal cord injury in primates}},
journal = {Scientific reports},
year = {2026},
month = may,
volume = {16},
number = {1},
pages = {24253},
publisher = {Nature Publishing Group},
issn = {2045-2322},
doi = {10.1038/s41598-026-54528-7},
url = {https://doi.org/10.1038/s41598-026-54528-7},
pmid = {42204263},
pmcid = {PMC13443211}
}

RIS

TY - JOUR
AU - Tohyama, Takamichi
AU - Yamaguchi, Reona
AU - Goda, Naokazu
AU - Yamamoto, Tetsuya
AU - Sadato, Norihiro
AU - Isa, Tadashi
AU - Fukunaga, Masaki
TI - Enhanced sensorimotor cortex responsiveness to nonplegic hand stimulation and motor network assembly during recovery after spinal cord injury in primates
T2 - Scientific reports
J2 - Sci Rep
PY - 2026
DA - 2026/05/27
VL - 16
IS - 1
SP - 24253
SN - 2045-2322
PB - Nature Publishing Group
DO - 10.1038/s41598-026-54528-7
UR - https://doi.org/10.1038/s41598-026-54528-7
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41598-026-54528-7",
"type": "article-journal",
"title": "Enhanced sensorimotor cortex responsiveness to nonplegic hand stimulation and motor network assembly during recovery after spinal cord injury in primates",
"container-title": "Scientific reports",
"author": [
{
"family": "Tohyama",
"given": "Takamichi"
},
{
"family": "Yamaguchi",
"given": "Reona"
},
{
"family": "Goda",
"given": "Naokazu"
},
{
"family": "Yamamoto",
"given": "Tetsuya"
},
{
"family": "Sadato",
"given": "Norihiro"
},
{
"family": "Isa",
"given": "Tadashi"
},
{
"family": "Fukunaga",
"given": "Masaki"
}
],
"container-title-short": "Sci Rep",
"volume": "16",
"issue": "1",
"page": "24253",
"DOI": "10.1038/s41598-026-54528-7",
"PMID": "42204263",
"PMCID": "PMC13443211",
"ISSN": "2045-2322",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41598-026-54528-7",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
27
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.7554/elife.108408 [code]
Frequency and laminar profile of feature-specific visual activity revealed by interleaved EEG-fMRI.
Journal: eLife
In common: DIPY, Tools for NIfTI and ANALYZE image (MATLAB), FSL, 6 other tools, fMRI, systems
[2] doi:10.1162/imag.a.1262 [code]
Frame-wise multi-echo distortion correction for superior functional MRI.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: pydicom, Tools for NIfTI and ANALYZE image (MATLAB), FSL, 6 other tools, fMRI
[3] doi:10.1038/s41586-026-10631-3 [code]
A prognostic human brain network for diffuse midline glioma.
Journal: Nature
In common: pydicom, Tools for NIfTI and ANALYZE image (MATLAB), FSL, 6 other tools, other condition
[4] doi:10.1002/mrm.70336 [code]
Offline Reconstruction of Diffusion MRI Acquisitions for Comparison Between Complex PCA-Based and AI-Based Denoising.
Journal: Magnetic resonance in medicine
In common: DIPY, Tools for NIfTI and ANALYZE image (MATLAB), FSL, 6 other tools
[5] doi:10.1111/nyas.70349 [code]
FREQ-NESS Reveals Age-Related Differences in Frequency-Resolved Brain Networks During Auditory Recognition and Resting State.
Journal: Annals of the New York Academy of Sciences
In common: Brain Connectivity Toolbox, Tools for NIfTI and ANALYZE image (MATLAB), FSL, 6 other tools
[6] doi:10.1038/s41398-026-04025-2 [code]
Brain energetic landscapes shape state dysregulation in major depressive disorder: a morphological network controllability perspective.
Journal: Translational psychiatry
In common: Brain Connectivity Toolbox, Tools for NIfTI and ANALYZE image (MATLAB), FSL, 6 other tools
[7] doi:10.1038/s41467-026-76011-7 [code]
Human cortex organizes dynamic co-fluctuations along the sensorimotor-association axis.
Journal: Nature communications
In common: Brain Connectivity Toolbox, FSL, SPM, 5 other tools, systems, 1 reference
[8] doi:10.1186/s13293-026-00891-z [code]
Sex differences in dynamic and static measures of brain integration derived from resting-state functional magnetic resonance imaging.
Journal: Biology of sex differences
In common: DIPY, pydicom, FSL, 5 other tools, fMRI
[9] doi:10.1038/s41467-026-73668-y [code]
Convergent and divergent brain-cognition development in early adolescence.
Journal: Nature communications
In common: Tools for NIfTI and ANALYZE image (MATLAB), FSL, SPM, 5 other tools, fMRI, 1 reference
[10] doi:10.1002/hbm.70483 [code]
Untamed: Unconstrained Tensor Decomposition and Graph Node Embedding for Cortical Parcellation.
Journal: Human brain mapping
In common: Tools for NIfTI and ANALYZE image (MATLAB), FSL, SPM, 5 other tools, fMRI, 1 reference

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.