OSCR

A portable solution for simultaneous human movement and mobile EEG acquisition: readiness potential for basketball free-throw shooting.

Code ↔ Paper

19 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 19 matches
  1. [1] § Materials and methods › EEG preprocessing ↔ scripts/A07_EEG_preprocessing.m, lines 52–159 · score 0.99 · pop_icflag, pop_iclabel, pop_viewprops, pop_interp, pop_reref, channel noise
  2. [2] § Materials and methods › EEG preprocessing ↔ scripts/A08_EEG_analysis_RP.m, lines 43–138 · score 0.93 · pop_interp, pop_reref, pop eegfiltnew, Bad components, baseline corrected, pre movement
  3. [3] § Materials and methods › Onset detection ↔ scripts/A03_ACC_sensor_rev_PRE.m, lines 314–397 · score 0.82 · reverse computation, artificial onset, baseline period, standard deviation, backwards, latency
  4. [4] § Materials and methods › Onset detection ↔ scripts/A04_ACC_sensor_rev_POST.m, lines 348–367 · score 0.80 · reverse computation, artificial onset, baseline period, standard deviation, latency, algorithm
  5. [5] § Results › Relationship between human pose and performance ↔ scripts/A17_MP_features.m, lines 31–139 · score 0.79 · left shoulder, right elbow, left pinky, right shoulder, right pinky, right wrist
  6. [6] § Results › Relationship between human pose and performance ↔ scripts/B01_MP_visualization.m, lines 149–246 · score 0.79 · left shoulder, right elbow, left pinky, right shoulder, right pinky, right wrist
  7. [7] § Results › Movement onset validation ↔ scripts/A26_RP_onset_imageplot_participant.m, lines 128–175 · score 0.78 · left thumb, right thumb, left pinky, right pinky, left wrist, right wrist
  8. [8] § Results › Movement onset validation ↔ scripts/A27_RP_onset_imageplot_grand_avg.m, lines 184–257 · score 0.78 · left thumb, right thumb, left pinky, right pinky, left wrist, right wrist
  9. [9] § Materials and methods › EEG preprocessing ↔ scripts/A07_EEG_preprocessing.m, lines 52–159 · score 0.77 · pop_jointprob, pop eegfiltnew, flagged, rejected, ICA, component
  10. [10] § Materials and methods › EEG preprocessing ↔ scripts/A06_EEG_preparation.m, lines 58–181 · score 0.77 · pop_jointprob, pop eegfiltnew, attenuation, ICA, filtered, root
  11. [11] § Materials and methods › Participants ↔ scripts/A32_Basketball_profile.m, lines 29–146 · score 0.65 · basketball profile, ages, week, clubs, contacts, activity
  12. [12] § Results › Movement onset validation ↔ scripts/A27_RP_onset_imageplot_grand_avg.m, lines 958–1080 · score 0.61 · RMS acceleration, acceleration magnitude, body landmarks, Body parts, stars, binomial
  13. [13] § Results › Presence of the RP ↔ scripts/A15_Topo_ttest_RP_avg.m, lines 245–360 · score 0.59 · Cz RP, channel Cz, FDR corrected, SE, Wilcoxon, 100 ms
  14. [14] § Materials and methods › Statistical analysis › Relationship between human pose and performance ↔ scripts/A19_Biserial_MP.m, lines 44–92 · score 0.57 · correlation coefficients, explained variance, vector, squaring, landmarks, PLD
  15. [15] § Materials and methods › Statistical analysis › Relationship between RP amplitudes and performance ↔ scripts/A18_Biserial_RP.m, lines 45–89 · score 0.56 · correlation coefficients, explained variance, vector, squaring, bin, RP
  16. [16] § Materials and methods › Statistical analysis › Relationship between RP amplitudes and performance ↔ scripts/A19_Biserial_MP.m, lines 95–164 · score 0.55 · correlation coefficients, explained variance, R2, squaring, amplitudes, hits
  17. [17] § Materials and methods › Materials ↔ scripts/B01_MP_visualization.m, lines 149–246 · score 0.55 · Pose Landmark Detection, MediaPipe, customized, timestamp, PLD, body
  18. [18] § Materials and methods › EEG data analysis ↔ scripts/A35_Interpolation_demo.m, lines 19–132 · score 0.53 · load_xdf, EEG timestamps, PLD
  19. [19] § Materials and methods › Statistical analysis › Movement onset validation ↔ scripts/A27_RP_onset_imageplot_grand_avg.m, lines 958–1080 · score 0.53 · acceleration magnitude, body landmarks, body part, Binomial, validation, onset

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 · 1,084 lines · 46 KB · MIT · 3 matches

  1. clc, clear, close all;
  3. %% Movement onset comparison (grand average)
  5. % This code compares the time reference (eye-wrist intersection) and the
  6. % grand average detected onset of the movement across participants.
  7. % Furthermore, a binomial test is used to determine whether the proportion
  8. % of participants showing significant movement in each body part was
  9. % significantly above chance level.
  11. % Miguel Contreras-Altamirano, 2025
  14. %% EEG data preparation
  16. mainpath = 'L:\Downloads\eeglab2023.0\'; % eeglab folder
  17. path = 'L:\Downloads\basketball_RP\NCP_Basketball\MediaPipe\';
  18. outpath = 'L:\Downloads\basketball_RP\Oldenburg_University\Thesis\data_hoops\';
  19. files = dir( fullfile( path,'\*.xdf')); % listing data sets
  21. % Add EEGLAB properly
  22. if exist('eeglab','file') ~= 2
  23. addpath(mainpath);
  24. end
  26. num_conditions = 3; % (Conditions and overall: 1=hit 2=miss 3=all)
  28. % Permutation parameters
  29. bin_size = 500; % Bin size in milliseconds
  30. num_permutations = 5000; % Number of permutations
  33. %% Loading data
  35. for cond=3 : num_conditions
  37. % Loading desired data
  38. if cond == 1 % 'hit'
  40. load([outpath , 'PLD_grand_avg_hit.mat'], 'averageTimeseriesMp', 'timestamps_mp');
  41. load([outpath , 'ACC_grand_avg_rev_hit','.mat']); % Loading accelerometer data
  43. % Import EEG processed data
  44. [ALLEEG EEG CURRENTSET ALLCOM] = eeglab; % Open eeglab
  45. EEG = pop_loadset('filename',['Grand_avg_hits','.set'],'filepath', outpath); % Loading set file
  48. elseif cond == 2 % 'miss'
  50. load([outpath , 'PLD_grand_avg_miss.mat'], 'averageTimeseriesMp', 'timestamps_mp');
  51. load([outpath , 'ACC_grand_avg_rev_miss','.mat']); % Loading accelerometer data
  53. % Import EEG processed data
  54. [ALLEEG EEG CURRENTSET ALLCOM] = eeglab; % Open eeglab
  55. EEG = pop_loadset('filename',['Grand_avg_misses','.set'],'filepath', outpath); % Loading set file
  56. eeglab redraw
  59. elseif cond == 3 % % 'none'
  61. load([outpath , 'PLD_grand_avg.mat'], 'averageTimeseriesMp', 'timestamps_mp');
  62. load([outpath , 'ACC_grand_avg_rev','.mat']); % Loading accelerometer data
  64. % Import EEG processed data
  65. [ALLEEG EEG CURRENTSET ALLCOM] = eeglab; % Open eeglab
  66. EEG = pop_loadset('filename',['Grand_avg_all','.set'],'filepath', outpath); % Loading set file
  67. eeglab redraw
  69. end
  71. %timeseries_mp = averageTimeseriesMp;
  74. frequency_sampling = input('At which frequency would you like the data? (1=15Hz / 2=250Hz): ');
  77. %% Reampling PLD to its originally 15 Hz
  79. if frequency_sampling == 1
  81. % Resampling instead
  82. %[OUTEEG] = pop_resample( INEEG, freq, fc, df);
  84. % Define original and target sampling rates
  85. original_sampling_rate = EEG.srate; % EEG sampling rate (e.g., 250Hz)
  86. target_sampling_rate = 15; % Target PLD sampling rate %sampling_rate_mp
  88. % Calculate downsampling factor
  89. downsample_factor = round(original_sampling_rate / target_sampling_rate);
  91. % Downsample the time series data
  92. averageTimeseriesMp = averageTimeseriesMp(:, 1:downsample_factor:end);
  94. timeseries_mp = averageTimeseriesMp;
  97. elseif frequency_sampling == 2
  99. timeseries_mp = averageTimeseriesMp;
  101. end
  104. %% ERP Onset with Gaussian
  106. % Assuming you have EEG data in an EEGLAB structure
  107. ERP_bin = mean (EEG.data, 3);
  108. chan = find(strcmp({EEG.chanlocs.labels}, 'Cz'));
  109. timeEEG = [find(EEG.times == -200)]; % for topoplot
  111. % Step 1: Extract the data for the channel of interest
  112. latency = EEG.times;
  113. erp_data = squeeze(EEG.data(chan, :, :)); % Squeeze the data to remove singleton dimensions and transpose
  114. sortvar = zeros(1, size(erp_data, 2)); % One value per trial (876 trials in this case)
  116. % Step 3: Use erpimage
  117. figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  118. %subplot(2, 1, 1);
  119. erpimage(erp_data, [], EEG.times, 'Grand Average ERP at Cz', 1, 1,...
  120. 'yerplabel', {'Grand Average Motion Tracking and ERP [N=26]'},'erp', 'off', 'cbar', 'off','cbar_title', '\muV',...
  121. 'vert', [grand_avgOnsetTime_rev], ...
  122. 'topo', {chan, EEG.chanlocs, EEG.chaninfo},...
  123. 'caxis', [-20 20],...
  124. 'vert', grand_avgOnsetTime_rev,... % 'avg_type', 'Gaussian',...
  125. 'img_trialax_label', {'Participants'}, 'img_trialax_ticks', [0 : 1 :size(erp_data,2)]);
  128. %% ERP Onset with Moving Average
  130. % Step 1: Define parameters
  131. clipping_range = [-20 20]; % Match the caxis range in erpimage
  133. % % Step 2: Apply smoothing
  134. % smooth_window = 20; % Adjust for slight smoothing if needed
  135. % smoothed_data = movmean(erp_data, smooth_window, 2); % Smooth across time
  137. smoothed_data = erp_data;
  139. % Step 3: Plot with imagesc
  140. figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  141. %subtightplot(2, 1, 2);
  142. imagesc(latency, 1:size(smoothed_data, 2), smoothed_data'); % Transpose for correct orientation
  143. set(gca, 'YDir', 'normal'); % Ensure correct orientation
  145. % Step 4: Colormap and axis adjustments
  146. colormap('jet'); % Similar to 'erpimage' colormap
  147. c = colorbar('Ticks',[-20 -10 0 10 20]); % Replace minValue and maxValue with your actual min and max
  148. c.Label.String = 'Amplitude [\muV]';
  149. c.Label.FontSize = 18;
  150. caxis(clipping_range); % Set color limits to match erpimage
  152. % Step 5: Add vertical lines and labels
  153. hold on;
  154. xline(0, 'r--', 'LineWidth', 2.5); % Onset line
  155. xline(grand_avgOnsetTime_rev, 'k:', 'LineWidth', 2.5); % Additional onset lines if needed
  156. xlabel('Time [ms]', 'FontSize', 18);
  157. ylabel('Participants', 'FontSize', 18);
  158. %title('Grand Average Motion Tracking and ERP [N=26]', 'FontSize', 20);
  159. %subtitle('Event Related Potential at [Cz]', 'FontSize', 19);
  161. % Optional: Adjust Y-axis tick labels
  162. yticks(1:size(smoothed_data, 2));
  163. yticklabels(1:size(smoothed_data, 2)); % Adjust as per participant indexing
  165. % % Step 4: Add the topoplot outside the matrix plot (above)
  166. % topoplot_axes = axes('Position', [0.56, 0.79, 0.1, 0.1]); % Adjust to position above the main plot
  167. % timeEEG = -200; % Choose the time point for topoplot (adjust as necessary)
  168. % topoplot(ERP_bin(:, find(latency == timeEEG)), EEG.chanlocs, 'maplimits', clipping_range, 'electrodes', 'off');
  169. % title([num2str(timeEEG), ' ms'], 'FontSize', 11);
  170. % set(topoplot_axes, 'Color', 'none'); % Optional: Make the topoplot background transparent
  171. %
  172. % % Step 4: Add the topoplot outside the matrix plot (above)
  173. % topoplot_axes = axes('Position', [0.50, 0.79, 0.1, 0.1]); % Adjust to position above the main plot
  174. % timeEEG = -500; % Choose the time point for topoplot (adjust as necessary)
  175. % topoplot(ERP_bin(:, find(latency == timeEEG)), EEG.chanlocs, 'maplimits', clipping_range, 'electrodes', 'off');
  176. % title([num2str(timeEEG), ' ms'], 'FontSize', 11);
  177. % set(topoplot_axes, 'Color', 'none'); % Optional: Make the topoplot background transparent
  179. % Save the Plot
  180. % saveas(gcf, [outpath, '\\group_analysis\\','ERP_onset_mov_grand_avg', '.jpg']); % Save the figure as a PNG image
  181. save_fig(gcf,[outpath, '\\group_analysis\\',], 'ERP_onset_mov_grand_avg', 'fontsize', 12);
  184. %% Calculate Acceleration Magnitude for Each Body Part
  186. % Assuming `averageTimeseriesMp` has dimensions [body_parts * 3 (xyz) * time]
  187. from = -2.5; % Start time in seconds
  188. to = 1; % End time in seconds
  190. % Reshape and initialize variables
  191. num_body_parts = 33;
  192. num_frames = size(averageTimeseriesMp, 2);
  193. acceleration_magnitude = zeros(num_body_parts, num_frames);
  195. % Calculate the acceleration magnitude for each body part
  196. for i = 1:num_body_parts
  197. % Extract x, y, and z rows for the current body part
  198. x = averageTimeseriesMp((i-1)*3 + 1, :);
  199. y = averageTimeseriesMp((i-1)*3 + 2, :);
  200. z = averageTimeseriesMp((i-1)*3 + 3, :); % --> Taking out z axis
  202. % Calculate magnitude
  203. acceleration_magnitude(i, :) = sqrt(x.^2 + y.^2 + z.^2); %
  204. end
  207. if frequency_sampling == 1
  209. % RMS over time for acceleration magnitude
  210. sec = 0.264; % Time window length in seconds (0.264 = 4 samples per window at 15Hz)
  211. LeWin = target_sampling_rate * sec; % Define window length in samples
  212. idx_loop = 1:max(1, LeWin):size(acceleration_magnitude, 2); % Ensure valid step size
  213. rms_acceleration_t = zeros(size(acceleration_magnitude, 1), length(idx_loop)); % Pre-allocate matrix
  215. % Loop through and calculate RMS of acceleration magnitude in each time window
  216. row_count = 1; % Initialize counter for rows
  217. for idx = 1:LeWin:size(acceleration_magnitude, 2) - LeWin
  218. signal = acceleration_magnitude(:, idx:idx + (LeWin - 1)); % Segment of acceleration magnitude
  219. rms_acceleration_t(:, row_count) = std(signal, [], 2); % Calculate RMS (standard deviation)
  220. row_count = row_count + 1;
  221. end
  224. elseif frequency_sampling == 2
  226. % RMS over time for acceleration magnitude
  227. sec = 0.01; % Time window length in seconds (0.01 = 2 samples per window at 250Hz)
  228. LeWin = EEG.srate * sec; % Define window length in samples
  229. idx_loop = 1:LeWin:size(acceleration_magnitude, 2); % Index for loop to go through windows
  230. rms_acceleration_t = zeros(size(acceleration_magnitude, 1), length(idx_loop)); % Pre-allocate matrix
  233. % Loop through and calculate RMS of acceleration magnitude in each time window
  234. row_count = 1; % Initialize counter for rows
  235. for idx = 1:LeWin:size(acceleration_magnitude, 2) - LeWin
  236. signal = acceleration_magnitude(:, idx:idx + (LeWin - 1)); % Segment of acceleration magnitude
  237. rms_acceleration_t(:, row_count) = std(signal, [], 2); % Calculate RMS (standard deviation)
  238. row_count = row_count + 1;
  239. end
  242. end
  246. % Multiply RMS values by 1000 to convert units if necessary
  247. rms_acceleration_t = rms_acceleration_t * 1000;
  249. % Define body part labels
  250. body_parts = {
  251. 'nose', 'left eye (inner)', 'left eye', 'left eye (outer)', 'right eye (inner)', ...
  252. 'right eye', 'right eye (outer)', 'left ear', 'right ear', 'mouth (left)', ...
  253. 'mouth (right)', 'left shoulder', 'right shoulder', 'left elbow', 'right elbow', ...
  254. 'left wrist', 'right wrist', 'left pinky', 'right pinky', 'left index', 'right index', ...
  255. 'left thumb', 'right thumb', 'left hip', 'right hip', 'left knee', 'right knee', ...
  256. 'left ankle', 'right ankle', 'left heel', 'right heel', 'left foot index', 'right foot index'
  257. };
  259. %%
  260. % %% Pose Landmarks Artifacts - 3D Surface Plot with Adjusted Labels for Y Coordinates
  261. %
  262. % % RMS measures overall variability, which is helpful for identifying general movement artifacts.
  263. % time_axis = linspace(from * 1000, to * 1000, length(idx_loop)); % in milliseconds
  264. %
  265. % % We will only label every third row (corresponding to the y-coordinates)
  266. % label_indices = 0:1:length(body_parts); % Indices for the y-coordinates (every third row)
  267. %
  268. % % Define time points for onset lines
  269. % onset_times = [0, grand_avgOnsetTime_rev]; % Replace with your actual onset times
  270. %
  271. % % Plot the 3D surface with correctly aligned Y-axis labels
  272. % figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  273. % [X, Y] = meshgrid(time_axis, 1:length(body_parts)); % Use body_parts length directly to match rows
  274. %
  275. % % Plot the surface
  276. % surf(X, Y, rms_acceleration_t, 'EdgeColor', 'none');
  277. % colormap("turbo");
  278. % c = colorbar;
  279. % c.Label.String = 'RMS of Acceleration Magnitude [mm/s^2]';
  280. % c.Label.FontSize = 11;
  281. % caxis([0 max(rms_acceleration_t(:))]);
  282. %
  283. % % Set axis limits and labels
  284. % xlim([min(time_axis) max(time_axis)]);
  285. % ylim([1 size(rms_acceleration_t, 1)]); % Ensure the Y-axis matches the data size
  286. %
  287. % % Adjust Y-axis labels
  288. % set(gca, 'YTick', 1:length(body_parts), 'YTickLabel', body_parts, 'YTickLabelRotation', 30, 'XTickLabelRotation', 30);
  289. %
  290. %
  291. % % Label axes
  292. % xlabel('Time [ms]', 'FontSize', 11);
  293. % ylabel('Body Landmarks', 'FontSize', 11);
  294. % zlabel('RMS', 'FontSize', 11);
  295. % title('Root Mean Square [RMS] of Acceleration Magnitude Over Time', 'FontSize', 12);
  296. % subtitle(['Windows of ', num2str(sec*1000), ' [ms]'], 'FontSize', 11.5);
  297. % view([110, 40]);
  298. %
  299. % % Recalculate and highlight top RMS body parts with correctly aligned labels
  300. % N = min(6, length(body_parts)); % Number of top body parts to highlight
  301. % [~, top_rms_indices] = maxk(mean(rms_acceleration_t, 2), N); % Get top body part indices by RMS
  302. % rms_threshold = min(max(rms_acceleration_t(label_indices(top_rms_indices), :), [], 2)); % Min of max RMS values for top body parts
  303. %
  304. %
  305. % % Draw 3D "walls" with restricted height at RMS threshold
  306. % hold on;
  307. % for i = 1:length(onset_times)
  308. % % Define the color and style for each line
  309. % if i == 1
  310. % line_color = 'r';
  311. % line_style = '--';
  312. % else
  313. % line_color = 'k';
  314. % line_style = '--';
  315. % end
  316. %
  317. % % Draw box edges at each onset time along x, y, and restricted z
  318. % plot3([onset_times(i), onset_times(i)], [1, 1], [0, rms_threshold], line_style, 'Color', line_color, 'LineWidth', 2); % Vertical edge
  319. % plot3([onset_times(i), onset_times(i)], [1, size(rms_acceleration_t, 1)], [0, 0], line_style, 'Color', line_color, 'LineWidth', 2); % Bottom horizontal edge
  320. % plot3([onset_times(i), onset_times(i)], [1, size(rms_acceleration_t, 1)], [rms_threshold, rms_threshold], line_style, 'Color', line_color, 'LineWidth', 2); % Top horizontal edge
  321. % end
  322. %
  323. % % Highlight top RMS body parts with labels
  324. % offset = 0.1; % Offset for label positioning
  325. % for i = 1:N
  326. % plot3(time_axis, repmat(top_rms_indices(i), size(time_axis)), rms_acceleration_t(top_rms_indices(i), :), 'b', 'Linestyle', ':', 'LineWidth', 2);
  327. % text(max(time_axis) + offset, top_rms_indices(i), max(rms_acceleration_t(top_rms_indices(i), :)), ...
  328. % body_parts{top_rms_indices(i)}, 'Color', 'k', 'FontSize', 8, 'FontWeight', 'bold', 'HorizontalAlignment', 'center');
  329. % end
  330. %
  331. % % Enable grid for better visualization
  332. % grid on;
  333. %
  334. % % Rotate the fig for better visualization
  335. % azimuth_angle = 3.349649635036496e+02;
  336. % elevation_angle = 81.093181818181820;
  337. %
  338. % view([azimuth_angle, elevation_angle]); % Replace with your desired values
  341. %% Plot Acceleration Magnitude as Coldmap 2D
  343. figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  344. imagesc(EEG.times, 1:size(rms_acceleration_t, 1), rms_acceleration_t);
  345. colorbar;
  346. colormap("sky");
  347. set(gca, 'YTick', 1:length(body_parts), 'YTickLabel', body_parts); % Label body parts
  348. xlabel('Time [ms]', 'FontSize', 11);
  349. ylabel('Body Landmarks', 'FontSize', 11);
  350. title('Root Mean Square [RMS] of Acceleration Magnitude', 'FontSize', 12);
  351. subtitle(['Windows of ', num2str(sec*1000), ' [ms]'], 'FontSize', 11.5);
  352. c = colorbar;
  353. c.Label.String = 'RMS of Acceleration Magnitude [mm/s^2]';
  354. c.Label.FontSize = 11;
  355. xline(0, '--', 'Color', 'r', 'LineWidth', 2); % Add red dashed line at 0 ms
  356. xline(grand_avgOnsetTime_rev, ':', 'Color', 'k', 'LineWidth', 2); % Add a red dashed line at 0 ms to indicate movement onset
  359. %% Plot Acceleration Magnitude as Heatmap 2D
  361. % % Plot with labels
  362. % fig_velocity = figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  363. % imagesc(EEG.times, 1:size(rms_acceleration_t, 1), rms_acceleration_t); % Plot RMS over time for MediaPipe data
  364. % colormap("turbo");
  365. % c = colorbar;
  366. % c.Label.String = 'RMS of Acceleration Magnitude [mm/s^2]';
  367. % c.Label.FontSize = 11;
  368. % set(gca, 'YTick', 1:1:33, 'YTickLabel', body_parts); % Adjust for your landmark indices
  369. % xlabel('Time [ms]', 'FontSize', 11);
  370. % ylabel('Body Landmarks', 'FontSize', 11);
  371. % title('Root Mean Square [RMS] of Acceleration Magnitude Over Time', 'FontSize', 12);
  372. % subtitle(['Windows of ', num2str(sec*1000), ' [ms]'], 'FontSize', 11.5);
  373. %
  374. % xline(0, '--', 'Color', 'r', 'LineWidth', 2); % Add a red dashed line at 0 ms to indicate movement onset
  375. % xline(grand_avgOnsetTime_rev, ':', 'Color', 'k', 'LineWidth', 2); % Add a red dashed line at 0 ms to indicate movement onset
  376. %
  377. % %saveas(gcf, [outpath, '\\group_analysis\\','RMS_motion_2D_grand_avg', '.jpg']); % Save the figure as a PNG image
  380. %% Permutation Test for Motion Significance
  382. % Parameters
  383. bin_samples = bin_size / 10; % Convert to samples (each sample is 10 ms)
  384. time_bins = -2500:bin_size:1000; % Define time bins in ms
  385. clusters = {'Head', 'Upper Body', 'Lower Body'};
  386. num_clusters = length(clusters);
  388. % Define cluster indices
  389. head_parts = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11];
  390. upper_body_parts = [12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23];
  391. lower_body_parts = [24, 25, 26, 27, 28, 29, 30, 31, 32, 33];
  393. cluster_parts = {head_parts, upper_body_parts, lower_body_parts};
  395. % Initialize results
  396. p_values = zeros(num_clusters, length(time_bins) - 1);
  399. % Perform Permutation Test for Each Cluster and Each Time Bin
  401. for cluster_idx = 1:length(clusters)
  402. cluster_data = rms_acceleration_t(cluster_parts{cluster_idx}, :); % Extract data for current cluster
  404. for bin_idx = 1:length(time_bins) - 1
  405. % Define start and end sample for the current bin
  406. start_sample = (bin_idx - 1) * bin_samples + 1;
  407. end_sample = min(start_sample + bin_samples - 1, size(cluster_data, 2));
  409. % Calculate the observed mean for the current bin
  410. observed_mean = mean(cluster_data(:, start_sample:end_sample), 'all');
  412. % Initialize permutation means
  413. permuted_means = zeros(1, num_permutations);
  415. % Permutation test
  416. for perm = 1:num_permutations
  417. % Permute data
  418. permuted_data = cluster_data(:, randperm(size(cluster_data, 2)));
  419. permuted_means(perm) = mean(permuted_data(:, start_sample:end_sample), 'all');
  420. end
  422. % Calculate p-value for observed mean
  423. p_values(cluster_idx, bin_idx) = sum(permuted_means >= observed_mean) / num_permutations;
  424. end
  425. end
  428. % Apply FDR Correction
  430. adjusted_p_values = mafdr(p_values(:), 'BHFDR', true);
  431. adjusted_p_values = reshape(adjusted_p_values, size(p_values));
  434. %% Visualization of Significant Time Bins
  436. figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  437. time_labels = time_bins(1:end-1) + bin_size / 2; % Midpoints of time bins
  439. % Define time bins from -2500 ms to 1000 ms
  440. bin_start_times = -2500:bin_size:1000; % Start of each 100 ms bin
  441. bin_labels = arrayfun(@(x) sprintf('%d %d ms', x), bin_start_times, 'UniformOutput', false); % Create labels
  443. for cluster_idx = 1:length(clusters)
  444. subplot(length(clusters), 1, cluster_idx);
  445. hold on;
  447. % Plot p-values
  448. plot(time_labels, adjusted_p_values(cluster_idx, :), '-o', 'LineWidth', 1.5);
  450. % Highlight significant bins
  451. significant_bins = adjusted_p_values(cluster_idx, :) < 0.05;
  452. scatter(time_labels(significant_bins), adjusted_p_values(cluster_idx, significant_bins), ...
  453. 50, 'r', 'filled'); % Red dots for significant bins
  455. % Labels and formatting
  456. title(['Cluster: ', clusters{cluster_idx}], 'FontSize', 14, 'FontWeight', 'bold');
  457. xlabel('Time [ms]', 'FontSize', 12);
  458. ylabel('Adjusted p-Value', 'FontSize', 12);
  459. ylim([0 1]);
  460. yline(0.05, '--', 'FDR Threshold (α = 0.05)', 'FontSize', 10, 'Color', 'k', 'LabelHorizontalAlignment','left');
  461. xline(0, 'r--', 'LineWidth', 2.5); % Onset line
  463. xlim([-2500, 1000]);
  464. xticks(bin_start_times); % Ensure all bins are represented
  465. xticklabels(bin_labels);
  466. %xtickangle(45); % Rotate labels for readability
  468. grid on;
  469. hold off;
  470. end
  472. sgtitle('Permutation Test for Motion Significance', 'FontSize', 16, 'FontWeight', 'bold');
  475. % Save the Plot
  476. %saveas(gcf, [outpath, '\\group_analysis\\', 'Permutation_test_motion_clusters', '.jpg']);
  477. % save_fig(gcf,[outpath, '\\group_analysis\\',], 'Permutation_test_motion_clusters');
  481. if frequency_sampling == 1
  483. saveas(gcf, [outpath, '\\group_analysis\\', 'Permutation_test_motion_clusters_15Hz', '.jpg']);
  485. elseif frequency_sampling == 2
  487. saveas(gcf, [outpath, '\\group_analysis\\', 'Permutation_test_motion_clusters_250Hz', '.jpg']);
  489. end
  493. %% Combined plots
  495. % Plot RMS Background
  496. figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  497. imagesc(EEG.times, 1:size(rms_acceleration_t, 1), rms_acceleration_t);
  498. colormap("turbo");
  499. c = colorbar;
  500. c.Label.String = 'RMS of Acceleration Magnitude [mm/s^2]';
  501. c.Label.FontSize = 11;
  502. set(gca, 'YTick', 1:1:33, 'YTickLabel', body_parts);
  503. xlabel('Time [ms]', 'FontSize', 11);
  504. ylabel('Body Landmarks', 'FontSize', 11, 'Position', [-2976.620246249664,16.560959874129882,1]);
  505. title('Root Mean Square [RMS] of Acceleration Magnitude Over Time', 'FontSize', 14);
  506. subtitle('Windows of 10 ms', 'FontSize', 11.5);
  508. % Add grid lines
  509. for c = [head_parts(end), upper_body_parts(end), lower_body_parts(end)]
  510. yline(c + 0.5, 'w-', 'LineWidth', 2);
  511. end
  514. hold on;
  515. % Draw vertical lines for time bins
  516. % for t = time_bins
  517. % xline(t, 'w--', 'LineWidth', 0.1);
  518. % end
  520. % Add cluster labels on the left side of the plot
  521. text(-2900, mean(head_parts), 'Head', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  522. text(-2900, mean(upper_body_parts)-2, 'Upper Body', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  523. text(-2900, mean(lower_body_parts)-2, 'Lower Body', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  526. % Highlight Significant Bins with Stars
  527. for cluster_idx = 1:num_clusters
  528. for bin_idx = 1:length(time_bins) - 1
  529. if adjusted_p_values(cluster_idx, bin_idx) < 0.05
  530. % Get the mid-time of the bin
  531. bin_mid_time = (time_bins(bin_idx) + time_bins(bin_idx + 1)) / 2;
  533. % Determine y-axis position for the star (center of the cluster range)
  534. y_center = mean(cluster_parts{cluster_idx});
  536. % Plot star
  537. plot(bin_mid_time, y_center, 'w*', 'MarkerSize', 8, 'LineWidth', 1.5);
  538. end
  539. end
  540. end
  542. xline(0, '--', 'Color', 'r', 'LineWidth', 2);
  543. xline(grand_avgOnsetTime_rev, ':', 'Color', 'k', 'LineWidth', 2); % Add a red dashed line at 0 ms to indicate movement onset
  544. hold off;
  547. % Save the Plot
  548. %saveas(gcf, [outpath, '\\group_analysis\\','RMS_motion_2D_grand_avg', '.jpg']); % Save the figure as a PNG image
  549. % save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_2D_grand_avg');
  554. if frequency_sampling == 1
  556. saveas(gcf, [outpath, '\\group_analysis\\','RMS_motion_2D_grand_avg_15Hz', '.jpg']); % Save the figure as a PNG image
  558. elseif frequency_sampling == 2
  560. saveas(gcf, [outpath, '\\group_analysis\\','RMS_motion_2D_grand_avg_250Hz', '.jpg']); % Save the figure as a PNG image
  562. end
  565. %% Combined plots 3D - Permutation
  567. % RMS measures overall variability, which is helpful for identifying general movement artifacts.
  568. time_axis = linspace(from * 1000, to * 1000, length(idx_loop)); % in milliseconds
  570. % We will only label every third row (corresponding to the y-coordinates)
  571. label_indices = 0:1:length(body_parts); % Indices for the y-coordinates (every third row)
  573. % Define time points for onset lines
  574. onset_times = [0, grand_avgOnsetTime_rev]; % Replace with your actual onset times
  576. % Plot the 3D surface with correctly aligned Y-axis labels
  577. figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  578. [X, Y] = meshgrid(time_axis, 1:length(body_parts)); % Use body_parts length directly to match rows
  580. % Plot the surface
  581. surf(X, Y, rms_acceleration_t, 'EdgeColor', 'none');
  582. colormap("turbo");
  583. c = colorbar;
  584. c.Label.String = 'RMS of Acceleration Magnitude [mm/s^2]';
  585. c.Label.FontSize = 11;
  586. caxis([0 max(rms_acceleration_t(:))]);
  588. % Set axis limits and labels
  589. xlim([min(time_axis) max(time_axis)]);
  590. ylim([1 size(rms_acceleration_t, 1)]); % Ensure the Y-axis matches the data size
  592. % Adjust Y-axis labels
  593. set(gca, 'YTick', 1:length(body_parts), 'YTickLabel', body_parts, 'XTickLabelRotation', 30); %'YTickLabelRotation', 10,
  595. % Label axes
  596. xlabel('Time [ms]', 'FontSize', 11);
  597. ylabel('Body Landmarks', 'FontSize', 11, 'Position', [-3438.001500791994,11.190612155547742,-0.16010789176363]);
  598. zlabel('RMS', 'FontSize', 11);
  599. %title('Root Mean Square [RMS] of Acceleration Magnitude Over Time', 'FontSize', 12);
  600. %subtitle(['Windows of ', num2str(sec*1000), ' [ms]'], 'FontSize', 11.5);
  601. view([110, 40]);
  603. % Recalculate and highlight top RMS body parts with correctly aligned labels
  604. N = min(6, length(body_parts)); % Number of top body parts to highlight
  605. [top_rms_values, top_rms_indices] = maxk(mean(rms_acceleration_t, 2), N); % Get top body part indices by RMS
  606. rms_threshold = min(max(rms_acceleration_t(label_indices(top_rms_indices), :), [], 2)); % Average of max RMS values for top body parts
  608. % Draw 3D "walls" with restricted height at RMS threshold
  609. hold on;
  610. for i = 1:length(onset_times)
  612. % Define the color and style for each line
  613. if i == 1
  614. line_color = 'r';
  615. line_style = '--';
  616. else
  617. line_color = 'k';
  618. line_style = ':';
  619. end
  621. % Draw box edges at each onset time along x, y, and restricted z
  622. plot3([onset_times(i), onset_times(i)], [1, 1], [0, rms_threshold], line_style, 'Color', line_color, 'LineWidth', 2.5); % Vertical edge
  623. plot3([onset_times(i), onset_times(i)], [1, size(rms_acceleration_t, 1)], [0, 0], line_style, 'Color', line_color, 'LineWidth', 2.5); % Bottom horizontal edge
  624. plot3([onset_times(i), onset_times(i)], [1, size(rms_acceleration_t, 1)], [rms_threshold, rms_threshold], line_style, 'Color', line_color, 'LineWidth', 2.5); % Top horizontal edge
  625. end
  627. % Highlight top RMS body parts with labels
  628. offset = 0.1; % Offset for label positioning
  629. for i = 1:N
  630. plot3(time_axis, repmat(top_rms_indices(i), size(time_axis)), rms_acceleration_t(top_rms_indices(i), :), 'b', 'Linestyle', ':', 'LineWidth', 2);
  631. %text(max(time_axis) + offset, top_rms_indices(i), max(rms_acceleration_t(top_rms_indices(i), :)), ...
  632. % body_parts{top_rms_indices(i)}, 'Color', 'k', 'FontSize', 8, 'HorizontalAlignment', 'center');
  633. end
  635. % Enable grid for better visualization
  636. grid on;
  638. % % Add grid lines
  639. % for c = [head_parts(end), upper_body_parts(end), lower_body_parts(end)]
  640. % yline(c + 0.5, 'w-', 'LineWidth', 2);
  641. % end
  643. hold on;
  644. % % Draw vertical lines for time bins
  645. % for t = time_bins
  646. % xline(t, 'w--', 'LineWidth', 0.1);
  647. % end
  649. % Add cluster labels on the left side of the plot
  650. text(-3200, mean(head_parts)+2, 'Head', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90); %, 'Rotation', 90
  651. text(-3100, mean(upper_body_parts)+2, 'Upper Body', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  652. text(-3100, mean(lower_body_parts)+1, 'Lower Body', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  654. % Determine maximum z value for positioning the stars on top
  655. max_rms_value = max(rms_acceleration_t(:));
  656. z_offset = 0.1; % Small offset to ensure the markers are clearly visible above the surface
  658. % Highlight Significant Bins with Stars on Top of the 3D Plot
  659. for cluster_idx = 1:num_clusters
  660. for bin_idx = 1:length(time_bins) - 1
  661. if adjusted_p_values(cluster_idx, bin_idx) < 0.05
  662. % Get the mid-time of the bin
  663. bin_mid_time = (time_bins(bin_idx) + time_bins(bin_idx + 1)) / 2;
  665. % Determine y-axis position for the star (center of the cluster range)
  666. y_center = mean(cluster_parts{cluster_idx});
  668. % Plot star at the top surface level
  669. plot3(bin_mid_time, y_center, max_rms_value + z_offset, ...
  670. 'Marker', "pentagram", 'Color', 'w', 'MarkerSize', 15, 'LineWidth', 1, 'MarkerEdgeColor', 'k', 'MarkerFaceColor','w'); % Place star slightly above the surface
  671. end
  672. end
  673. end
  675. % Rotate the fig for better visualization
  676. azimuth_angle = -20.695036496350380;
  677. elevation_angle = 82.462139689578750;
  678. view([azimuth_angle, elevation_angle]); % Replace with your desired values
  681. % Loop through significant bins and highlight them with transparent 3D planes
  682. for cluster_idx = 1:num_clusters
  683. for bin_idx = 1:length(time_bins) - 1
  684. if adjusted_p_values(cluster_idx, bin_idx) < 0.05
  685. % Get the time range of the bin
  686. bin_start_time = time_bins(bin_idx);
  687. bin_end_time = time_bins(bin_idx + 1);
  689. % Determine y-axis range for the cluster
  690. y_start = min(cluster_parts{cluster_idx});
  691. y_end = max(cluster_parts{cluster_idx});
  693. % Define z-axis range (RMS values) for the box
  694. z_start = 0;
  695. z_end = max_rms_value;
  697. % Create vertices for the 3D box
  698. vertices = [
  699. bin_start_time, y_start, z_start; % Bottom-front-left
  700. bin_start_time, y_end, z_start; % Bottom-front-right
  701. bin_end_time, y_end, z_start; % Bottom-back-right
  702. bin_end_time, y_start, z_start; % Bottom-back-left
  703. bin_start_time, y_start, z_end; % Top-front-left
  704. bin_start_time, y_end, z_end; % Top-front-right
  705. bin_end_time, y_end, z_end; % Top-back-right
  706. bin_end_time, y_start, z_end; % Top-back-left
  707. ];
  709. % Define faces of the box
  710. faces = [
  711. 1, 2, 6, 5; % Front
  712. 2, 3, 7, 6; % Right
  713. 3, 4, 8, 7; % Back
  714. 4, 1, 5, 8; % Left
  715. 5, 6, 7, 8; % Top
  716. 1, 2, 3, 4; % Bottom
  717. ];
  719. % Add the 3D box to the plot
  720. patch('Vertices', vertices, 'Faces', faces, ...
  721. 'FaceColor', "yellow", 'FaceAlpha', 0.08, 'EdgeColor', 'k');
  722. end
  723. end
  724. end
  726. %saveas(gcf, [outpath, '\\group_analysis\\','RMS_motion_3D_grand_avg', '.jpg']); % Save the figure as a PNG image
  727. %save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_3D_grand_avg_Permutation', 'fontsize', 10);
  729. % Save a figure with 2-column width (180 mm) as TIFF
  730. % save_fig_pro(gcf, [outpath, '\\group_analysis\\',], 'RMS_motion_3D_grand_avg', 'fontsize', 8, 'width_mm', 180, 'figtype', 'tiff', 'dpi', 600);
  736. if frequency_sampling == 1
  738. save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_3D_grand_avg_Permutation_15Hz', 'fontsize', 10);
  740. elseif frequency_sampling == 2
  742. save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_3D_grand_avg_Permutation_250Hz', 'fontsize', 10);
  744. end
  748. %% Combined plots 3D - Binomial
  750. % RMS measures overall variability, which is helpful for identifying general movement artifacts.
  751. time_axis = linspace(from * 1000, to * 1000, length(idx_loop)); % in milliseconds
  753. % We will only label every third row (corresponding to the y-coordinates)
  754. label_indices = 0:1:length(body_parts); % Indices for the y-coordinates (every third row)
  756. % Define time points for onset lines
  757. onset_times = [0, grand_avgOnsetTime_rev]; % Replace with your actual onset times
  759. % Plot the 3D surface with correctly aligned Y-axis labels
  760. figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  761. [X, Y] = meshgrid(time_axis, 1:length(body_parts)); % Use body_parts length directly to match rows
  763. % Plot the surface
  764. surf(X, Y, rms_acceleration_t, 'EdgeColor', 'none');
  765. colormap("turbo");
  766. c = colorbar;
  767. c.Label.String = 'RMS of Acceleration Magnitude [mm/s^2]';
  768. c.Label.FontSize = 11;
  769. caxis([0 max(rms_acceleration_t(:))]);
  771. % Set axis limits and labels
  772. xlim([min(time_axis) max(time_axis)]);
  773. ylim([1 size(rms_acceleration_t, 1)]); % Ensure the Y-axis matches the data size
  775. % Adjust Y-axis labels
  776. set(gca, 'YTick', 1:length(body_parts), 'YTickLabel', body_parts, 'XTickLabelRotation', 30); %'YTickLabelRotation', 10,
  778. % Label axes
  779. xlabel('Time [ms]', 'FontSize', 11);
  780. ylabel('Body Landmarks', 'FontSize', 11, 'Position', [-3438.001500791994,11.190612155547742,-0.16010789176363]);
  781. zlabel('RMS', 'FontSize', 11);
  782. %title('Root Mean Square [RMS] of Acceleration Magnitude Over Time', 'FontSize', 12);
  783. %subtitle(['Windows of ', num2str(sec*1000), ' [ms]'], 'FontSize', 11.5);
  784. view([110, 40]);
  786. % Recalculate and highlight top RMS body parts with correctly aligned labels
  787. N = min(6, length(body_parts)); % Number of top body parts to highlight
  788. [top_rms_values, top_rms_indices] = maxk(mean(rms_acceleration_t, 2), N); % Get top body part indices by RMS
  789. rms_threshold = min(max(rms_acceleration_t(label_indices(top_rms_indices), :), [], 2)); % Average of max RMS values for top body parts
  791. % Draw 3D "walls" with restricted height at RMS threshold
  792. hold on;
  793. for i = 1:length(onset_times)
  795. % Define the color and style for each line
  796. if i == 1
  797. line_color = 'r';
  798. line_style = '--';
  799. else
  800. line_color = 'k';
  801. line_style = ':';
  802. end
  804. % Draw box edges at each onset time along x, y, and restricted z
  805. plot3([onset_times(i), onset_times(i)], [1, 1], [0, rms_threshold], line_style, 'Color', line_color, 'LineWidth', 2.5); % Vertical edge
  806. %plot3([onset_times(i), onset_times(i)], [1, size(rms_acceleration_t, 1)], [0, 0], line_style, 'Color', line_color, 'LineWidth', 2.5); % Bottom horizontal edge
  807. plot3([onset_times(i), onset_times(i)], [1, size(rms_acceleration_t, 1)], [rms_threshold, rms_threshold], line_style, 'Color', line_color, 'LineWidth', 2.5); % Top horizontal edge
  808. end
  810. % % Highlight top RMS body parts with labels
  811. % offset = 0.1; % Offset for label positioning
  812. % for i = 1:N
  813. % plot3(time_axis, repmat(top_rms_indices(i), size(time_axis)), rms_acceleration_t(top_rms_indices(i), :), 'b', 'Linestyle', ':', 'LineWidth', 2);
  814. % %text(max(time_axis) + offset, top_rms_indices(i), max(rms_acceleration_t(top_rms_indices(i), :)), ...
  815. % % body_parts{top_rms_indices(i)}, 'Color', 'k', 'FontSize', 8, 'HorizontalAlignment', 'center');
  816. % end
  818. % Enable grid for better visualization
  819. grid on;
  821. % % Add grid lines
  822. % for c = [head_parts(end), upper_body_parts(end), lower_body_parts(end)]
  823. % yline(c + 0.5, 'w-', 'LineWidth', 2);
  824. % end
  826. hold on;
  827. % % Draw vertical lines for time bins
  828. % for t = time_bins
  829. % xline(t, 'w--', 'LineWidth', 0.1);
  830. % end
  832. % Add cluster labels on the left side of the plot
  833. text(-3200, mean(head_parts)+2, 'Head', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90); %, 'Rotation', 90
  834. text(-3100, mean(upper_body_parts)+2, 'Upper Body', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  835. text(-3100, mean(lower_body_parts)+1, 'Lower Body', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  837. % % Determine maximum z value for positioning the stars on top
  838. % max_rms_value = max(rms_acceleration_t(:));
  839. % z_offset = 0.1; % Small offset to ensure the markers are clearly visible above the surface
  840. %
  841. % % Highlight Significant Bins with Stars on Top of the 3D Plot
  842. % for cluster_idx = 1:num_clusters
  843. % for bin_idx = 1:length(time_bins) - 1
  844. % if adjusted_p_values(cluster_idx, bin_idx) < 0.05
  845. % % Get the mid-time of the bin
  846. % bin_mid_time = (time_bins(bin_idx) + time_bins(bin_idx + 1)) / 2;
  847. %
  848. % % Determine y-axis position for the star (center of the cluster range)
  849. % y_center = mean(cluster_parts{cluster_idx});
  850. %
  851. % % Plot star at the top surface level
  852. % plot3(bin_mid_time, y_center, max_rms_value + z_offset, ...
  853. % 'Marker', "pentagram", 'Color', 'w', 'MarkerSize', 15, 'LineWidth', 1, 'MarkerEdgeColor', 'k', 'MarkerFaceColor','w'); % Place star slightly above the surface
  854. % end
  855. % end
  856. % end
  858. % Rotate the fig for better visualization
  859. azimuth_angle = -20.695036496350380;
  860. elevation_angle = 82.462139689578750;
  861. view([azimuth_angle, elevation_angle]); % Replace with your desired values
  863. % % Loop through significant bins and highlight them with transparent 3D planes
  864. % for cluster_idx = 1:num_clusters
  865. % for bin_idx = 1:length(time_bins) - 1
  866. % if adjusted_p_values(cluster_idx, bin_idx) < 0.05
  867. % % Get the time range of the bin
  868. % bin_start_time = time_bins(bin_idx);
  869. % bin_end_time = time_bins(bin_idx + 1);
  870. %
  871. % % Determine y-axis range for the cluster
  872. % y_start = min(cluster_parts{cluster_idx});
  873. % y_end = max(cluster_parts{cluster_idx});
  874. %
  875. % % Define z-axis range (RMS values) for the box
  876. % z_start = 0;
  877. % z_end = max_rms_value;
  878. %
  879. % % Create vertices for the 3D box
  880. % vertices = [
  881. % bin_start_time, y_start, z_start; % Bottom-front-left
  882. % bin_start_time, y_end, z_start; % Bottom-front-right
  883. % bin_end_time, y_end, z_start; % Bottom-back-right
  884. % bin_end_time, y_start, z_start; % Bottom-back-left
  885. % bin_start_time, y_start, z_end; % Top-front-left
  886. % bin_start_time, y_end, z_end; % Top-front-right
  887. % bin_end_time, y_end, z_end; % Top-back-right
  888. % bin_end_time, y_start, z_end; % Top-back-left
  889. % ];
  890. %
  891. % % Define faces of the box
  892. % faces = [
  893. % 1, 2, 6, 5; % Front
  894. % 2, 3, 7, 6; % Right
  895. % 3, 4, 8, 7; % Back
  896. % 4, 1, 5, 8; % Left
  897. % 5, 6, 7, 8; % Top
  898. % 1, 2, 3, 4; % Bottom
  899. % ];
  900. %
  901. % % Add the 3D box to the plot
  902. % patch('Vertices', vertices, 'Faces', faces, ...
  903. % 'FaceColor', "yellow", 'FaceAlpha', 0.08, 'EdgeColor', 'k');
  904. % end
  905. % end
  906. % end
  909. % Add Significant Movement Stars to the Plot
  911. % Define the path to the onset validation file
  912. onset_validation_file = fullfile(outpath, 'binomial_test_acceleration.xlsx');
  914. % Read the onset validation data
  915. onset_data = readtable(onset_validation_file, 'ReadVariableNames', true);
  917. % Identify significant body parts based on p-value threshold
  918. significant_parts = onset_data.pValue < 0.05;
  919. significant_body_parts = onset_data.parts(significant_parts);
  921. % Loop through the body parts and find their indices
  922. significant_indices = [];
  923. for i = 1:length(significant_body_parts)
  924. % Find the index of the significant body part in the body_parts list
  925. idx = find(strcmp(body_parts, significant_body_parts{i}));
  926. if ~isempty(idx)
  927. significant_indices = [significant_indices; idx];
  928. end
  929. end
  931. % Add stars for significant body parts at time 0
  932. hold on;
  933. for i = 1:length(significant_indices)
  934. idx = significant_indices(i);
  935. plot3(0, idx, max(rms_acceleration_t(:)) + 0.1, 'p', ...
  936. 'Marker', "pentagram", 'Color', 'w', 'MarkerSize', 15, 'LineWidth', 1, 'MarkerEdgeColor', 'k', 'MarkerFaceColor','w');
  937. end
  939. %saveas(gcf, [outpath, '\\group_analysis\\','RMS_motion_3D_grand_avg_PRO', '.jpg']); % Save the figure as a PNG image
  940. %save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_3D_grand_avg_Binomial', 'fontsize', 10);
  944. if frequency_sampling == 1
  946. save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_3D_grand_avg_Binomial_15Hz', 'fontsize', 10);
  948. elseif frequency_sampling == 2
  950. save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_3D_grand_avg_Binomial_250Hz', 'fontsize', 10);
  952. end
  955. disp('Updated plot with significant body parts saved!');
  958. %% Combined plots 2D - Binomial
  960. % RMS measures overall variability, which is helpful for identifying general movement artifacts.
  961. time_axis = linspace(from * 1000, to * 1000, length(idx_loop)); % in milliseconds
  963. % We will only label every third row (corresponding to the y-coordinates)
  964. label_indices = 0:1:length(body_parts); % Indices for the y-coordinates (every third row)
  966. % Define time points for onset lines
  967. onset_times = [0, grand_avgOnsetTime_rev]; % Replace with your actual onset times
  969. % Plot the 2D surface with correctly aligned Y-axis labels
  970. figure('units', 'normalized', 'outerposition', [0 0 1 1]);
  971. imagesc(time_axis, 1:size(rms_acceleration_t, 1), rms_acceleration_t);
  972. colormap("turbo");
  973. c = colorbar;
  974. c.Label.String = 'RMS of Acceleration Magnitude [mm/s^2]';
  975. c.Label.FontSize = 11;
  976. caxis([0 max(rms_acceleration_t(:))]);
  977. set(gca, 'YTick', 1:1:33, 'YTickLabel', body_parts);
  978. xlabel('Time [ms]', 'FontSize', 11);
  979. ylabel('Body Landmarks', 'FontSize', 11, 'Position', [-2976.620246249664,16.560959874129882,1]);
  980. % title('Root Mean Square [RMS] of Acceleration Magnitude Over Time', 'FontSize', 14);
  981. % subtitle('Windows of 10 ms', 'FontSize', 11.5);
  983. % Set axis limits and labels
  984. xlim([min(time_axis) max(time_axis)]);
  985. ylim([1 size(rms_acceleration_t, 1)]); % Ensure the Y-axis matches the data size
  987. % % Add grid lines
  988. % for c = [head_parts(end), upper_body_parts(end), lower_body_parts(end)]
  989. % yline(c + 0.5, 'w-', 'LineWidth', 2);
  990. % end
  992. hold on;
  994. % Add cluster labels on the left side of the plot
  995. text(-2900, mean(head_parts), 'Head', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  996. text(-2900, mean(upper_body_parts)-2, 'Upper Body', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  997. text(-2900, mean(lower_body_parts)-2, 'Lower Body', 'FontSize', 11, 'FontWeight', 'bold', 'HorizontalAlignment', 'right', 'Rotation', 90);
  1000. % Recalculate and highlight top RMS body parts with correctly aligned labels
  1001. N = min(6, length(body_parts)); % Number of top body parts to highlight
  1002. [top_rms_values, top_rms_indices] = maxk(mean(rms_acceleration_t, 2), N); % Get top body part indices by RMS
  1003. rms_threshold = min(max(rms_acceleration_t(label_indices(top_rms_indices), :), [], 2)); % Average of max RMS values for top body parts
  1005. % Draw 3D "walls" with restricted height at RMS threshold
  1006. hold on;
  1007. for i = 1:length(onset_times)
  1009. % Define the color and style for each line
  1010. if i == 1
  1011. line_color = 'r';
  1012. line_style = '--';
  1013. else
  1014. line_color = 'k';
  1015. line_style = ':';
  1016. end
  1018. % Draw box edges at each onset time along x, y, and restricted z
  1019. plot3([onset_times(i), onset_times(i)], [1, 1], [0, rms_threshold], line_style, 'Color', line_color, 'LineWidth', 2.5); % Vertical edge
  1020. %plot3([onset_times(i), onset_times(i)], [1, size(rms_acceleration_t, 1)], [0, 0], line_style, 'Color', line_color, 'LineWidth', 2.5); % Bottom horizontal edge
  1021. plot3([onset_times(i), onset_times(i)], [1, size(rms_acceleration_t, 1)], [rms_threshold, rms_threshold], line_style, 'Color', line_color, 'LineWidth', 2.5); % Top horizontal edge
  1022. end
  1024. % Add Significant Movement Stars to the Plot
  1025. hold on;
  1027. % Define the path to the onset validation file
  1028. onset_validation_file = fullfile(outpath, 'binomial_test_acceleration.xlsx');
  1030. % Read the onset validation data
  1031. onset_data = readtable(onset_validation_file, 'ReadVariableNames', true);
  1033. % Identify significant body parts based on p-value threshold
  1034. significant_parts = onset_data.pValue < 0.05;
  1035. significant_body_parts = onset_data.parts(significant_parts);
  1037. % Loop through the body parts and find their indices
  1038. significant_indices = [];
  1039. for i = 1:length(significant_body_parts)
  1040. % Find the index of the significant body part in the body_parts list
  1041. idx = find(strcmp(body_parts, significant_body_parts{i}));
  1042. if ~isempty(idx)
  1043. significant_indices = [significant_indices; idx];
  1044. end
  1045. end
  1047. % Add stars for significant body parts at time 0
  1048. hold on;
  1049. for i = 1:length(significant_indices)
  1050. idx = significant_indices(i);
  1051. plot3(0, idx, max(rms_acceleration_t(:)) + 0.1, 'p', ...
  1052. 'Marker', "pentagram", 'Color', 'w', 'MarkerSize', 15, 'LineWidth', 1, 'MarkerEdgeColor', 'k', 'MarkerFaceColor','w');
  1053. end
  1055. %saveas(gcf, [outpath, '\\group_analysis\\','RMS_motion_3D_grand_avg_PRO', '.jpg']); % Save the figure as a PNG image
  1056. %save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_2D_grand_avg_Binomial', 'fontsize', 10);
  1060. if frequency_sampling == 1
  1062. save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_2D_grand_avg_Binomial_15Hz', 'fontsize', 10);
  1064. acceleration_magnitude_15Hz = rms_acceleration_t;
  1065. time_axis_15Hz = time_axis;
  1067. save([outpath, 'Acceleration_magnitude_grand_avg_15Hz', '.mat'],'acceleration_magnitude_15Hz', 'time_axis_15Hz');
  1069. elseif frequency_sampling == 2
  1071. save_fig(gcf,[outpath, '\\group_analysis\\',], 'RMS_motion_2D_grand_avg_Binomial_250Hz', 'fontsize', 10);
  1073. acceleration_magnitude_250Hz = rms_acceleration_t;
  1074. time_axis_250Hz = time_axis;
  1076. save([outpath, 'Acceleration_magnitude_grand_avg_250Hz', '.mat'],'acceleration_magnitude_250Hz', 'time_axis_250Hz');
  1078. end
  1080. end

A27_RP_onset_imageplot_grand_avg.m at commit 8753e5e, under MIT · at the source

Overview

Authors: Miguel Contreras-Altamirano1, Melanie Klapprott1, Nadine Jacobsen1, Paul Maanen1,2, Julius Welzel3, Stefan Debener1,2,4,5
ORCID iDs: Miguel Contreras-Altamirano
  1. Neuropsychology Lab, Department of Psychology, School of Medicine and Health Sciences, Carl Von Ossietzky Universität Oldenburg, Oldenburg, Germany
  2. Cluster of Excellence “Hearing4All”, Carl Von Ossietzky Universität Oldenburg, Oldenburg, Germany
  3. Kiel University, Kiel, Germany
  4. Fraunhofer Institute of Digital Media Technology, Oldenburg Branch for Hearing, Oldenburg, Germany
  5. Center for Neurosensory Science and Systems, Carl Von Ossietzky University of Oldenburg, Oldenburg, Germany
Institutions: Carl von Ossietzky Universität Oldenburg (Germany); Hearing4all (Germany); Christian-Albrechts-Universität zu Kiel (Germany); Fraunhofer Institute for Digital Media Technology (Germany)
Journal: Experimental brain research, volume 244, issue 8, article 153
Dates: received 24 March 2026; accepted 12 June 2026; published online 7 July 2026; in print 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1007/s00221-026-07342-6 · PMID 42412206 · PMCID PMC13342168 · OpenAlex W4406273067
Open access: hybrid, a free copy (OpenAlex)
Preprint: arxiv:2501.05378
Status: code verified
Categories: EEG (modality), human (organism)
Methods: Connectivity, Statistics, Smoothing, state filtering, decompositions, Preprocessing, Evoked potentials, Machine learning
Keywords: Mobile EEG, Human pose, Readiness potential, MoBI, Basketball
MeSH: Basketball*, Contingent Negative Variation*, Electroencephalography*, Motor Activity*, Psychomotor Performance*, Smartphone*, Adult, Female, Humans, Male, Movement, Young Adult (* major topic)
Topic: EEG and Brain-Computer Interfaces (Cognitive Neuroscience, Neuroscience), according to OpenAlex
Funding: Carl von Ossietzky Universität Oldenburg
Citations: not cited yet (Europe PMC); 145 references in the paper
Research resources: RRID:SCR_013726

Abstract

Advances in wireless electroencephalography (EEG) technology promise to record brain-electrical activity in everyday situations. To better understand the relationship between brain activity and natural behavior, it is necessary to monitor human movement patterns. Here, we present a pocketable setup consisting of two smartphones to simultaneously capture human posture and EEG signals. We asked 26 basketball players to shoot 120 free throws each. First, we investigated whether our setup allows us to capture the readiness potential (RP) that precedes voluntary actions. Second, we investigated whether the RP differs between successful and unsuccessful free-throw attempts. The results confirmed the presence of the RP over fronto-central channels, with significant negative deflection at channel Cz, from − 400 to 0 ms before movement onset (M ± SE: − 6.54 ± 2.26 to − 13.52 ± 2.42 μV; z = − 2.53 to − 3.92; FDR-corrected p = 0.049 to 0.003; r = 0.50 to 0.77). However, the amplitude of the RP was not related to shooting success (all FDR-corrected p > 0.05; maximum mean R2 = 0.047, i.e., 4.7% explained variance). Preliminary exploratory pose analysis conducted offline indicated the presence of participant-specific variations in posture between successful and unsuccessful shots in 38.5% of participants (10/26), with 4.5% explained variance (maximum mean landmark R2 = 0.045). We conclude that a highly portable, low-cost and lightweight acquisition setup, consisting of two smartphones and a head-mounted wireless EEG amplifier, is sufficient to monitor complex human movement patterns and associated brain dynamics outside the laboratory.

Supplementary Information: The online version contains supplementary material available at 10.1007/s00221-026-07342-6.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 19 matches between paragraphs and lines of code.

micufx/Pocketable-MoBI-Baskts

License: MIT
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: 8753e5efbceea1d510fc1fa9c22ab3f4a0da19f5, 5 August 2026
Languages: MATLAB (48)
Size: 80 files, 48 scripts
Software Heritage: not archived
Found in: “Data availability”
Holds: README, environment (pyproject.toml)
Not found: license file, CITATION.cff, tests, continuous integration, documentation
Tools: EEGLAB (23 files), Statistics and Machine Learning Toolbox (8 files), FieldTrip (1 file), ICLabel (1 file), Signal Processing Toolbox (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
49 files
At the source: github.com/micufx/Pocketable-MoBI-Baskts

The paper's code and data availability statement is in the Data section.

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;
  • 48 scripts, each with its path and the digest of its content;
  • 19 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

The data that support the findings of this study are available on request from the corresponding author. The MATLAB code is available on GitHub (https://github.com/micufx/Pocketable-MoBI-Baskts). The smartphone LSL apps Senda and Recorda are available on GitHub (https://neuropsyol.github.io/). All information about the set-up is available at https://juliuswelzel.github.io/eeg_basketball_website/.

Reproduced under the paper's license (CC BY), from the paper cited above.

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 2, 28 September 2026

  • Publisher: n/a → Springer Science+Business Media

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 6 authors, 5 keywords, 12 MeSH terms, 1 funder, 129 references, 1 RRID.

Cite

This paper

Contreras-Altamirano, M., Klapprott, M., Jacobsen, N., Maanen, P., Welzel, J., & Debener, S. (2026). A portable solution for simultaneous human movement and mobile EEG acquisition: readiness potential for basketball free-throw shooting. Experimental brain research, 244(8), 153. https://doi.org/10.1007/s00221-026-07342-6

BibTeX

@article{contrerasaltamirano2026portable,
author = {Contreras-Altamirano, Miguel and Klapprott, Melanie and Jacobsen, Nadine and Maanen, Paul and Welzel, Julius and Debener, Stefan},
title = {{A portable solution for simultaneous human movement and mobile EEG acquisition: readiness potential for basketball free-throw shooting}},
journal = {Experimental brain research},
year = {2026},
month = jul,
volume = {244},
number = {8},
pages = {153},
publisher = {Springer Science+Business Media},
issn = {0014-4819},
doi = {10.1007/s00221-026-07342-6},
url = {https://doi.org/10.1007/s00221-026-07342-6},
pmid = {42412206},
pmcid = {PMC13342168}
}

RIS

TY - JOUR
AU - Contreras-Altamirano, Miguel
AU - Klapprott, Melanie
AU - Jacobsen, Nadine
AU - Maanen, Paul
AU - Welzel, Julius
AU - Debener, Stefan
TI - A portable solution for simultaneous human movement and mobile EEG acquisition: readiness potential for basketball free-throw shooting
T2 - Experimental brain research
J2 - Exp Brain Res
PY - 2026
DA - 2026/07/07
VL - 244
IS - 8
SP - 153
SN - 0014-4819
PB - Springer Science+Business Media
DO - 10.1007/s00221-026-07342-6
UR - https://doi.org/10.1007/s00221-026-07342-6
LA - en
ER -

CSL-JSON

{
"id": "10.1007/s00221-026-07342-6",
"type": "article-journal",
"title": "A portable solution for simultaneous human movement and mobile EEG acquisition: readiness potential for basketball free-throw shooting",
"container-title": "Experimental brain research",
"author": [
{
"family": "Contreras-Altamirano",
"given": "Miguel"
},
{
"family": "Klapprott",
"given": "Melanie"
},
{
"family": "Jacobsen",
"given": "Nadine"
},
{
"family": "Maanen",
"given": "Paul"
},
{
"family": "Welzel",
"given": "Julius"
},
{
"family": "Debener",
"given": "Stefan"
}
],
"container-title-short": "Exp Brain Res",
"volume": "244",
"issue": "8",
"page": "153",
"DOI": "10.1007/s00221-026-07342-6",
"PMID": "42412206",
"PMCID": "PMC13342168",
"ISSN": "0014-4819",
"publisher": "Springer Science+Business Media",
"URL": "https://doi.org/10.1007/s00221-026-07342-6",
"language": "en",
"issued": {
"date-parts": [
[
2026,
7,
7
]
]
}
}

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:10.1016/j.isci.2026.116647 [code]
Enhancing mobile brain and body imaging: Open-source solutions for real-world research applications.
Journal: iScience
In common: 14 references, author Miguel Contreras-Altamirano
[2] doi:10.1111/psyp.70352 [code]
Preprocessing on the Go: Practices in Gait-Related Mobile EEG.
Journal: Psychophysiology
In common: EEG, 10 references
[3] doi:10.3390/s26082440 [code]
Suppressing Non-Stationary Motion Artefacts in Mobile EEG Using Generalized Eigenvalue Decomposition.
Journal: Sensors (Basel, Switzerland)
In common: EEG, 7 references
[4] doi:10.7554/elife.107088 [code]
Development of auditory and spontaneous movement responses to music over the first postnatal year.
Journal: eLife
In common: ICLabel, EEGLAB, FieldTrip, 2 other tools, EEG, 2 references
[5] doi:10.1162/imag.a.1201 [code]
All spectral frequencies of neural activity reveal semantic representation in the human anterior ventral temporal cortex.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: ICLabel, EEGLAB, FieldTrip, 2 other tools, 2 references
[6] doi:10.1093/pnasnexus/pgag220 [code]
Neural mechanisms of awareness of action.
Journal: PNAS nexus
In common: EEGLAB, Signal Processing Toolbox, Statistics and Machine Learning Toolbox, EEG, 3 references
[7] doi:10.1038/s41598-026-56070-y [code]
SSDLabeler: realistic semi-synthetic data generation for multi-label artifact classification in EEG.
Journal: Scientific reports
In common: ICLabel, EEGLAB, Signal Processing Toolbox, 1 other tool, EEG, 2 references
[8] doi:10.1523/eneuro.0316-25.2026 [code]
Neural Mechanisms of Self-Generated Action Sequences.
Journal: eNeuro
In common: EEGLAB, FieldTrip, Statistics and Machine Learning Toolbox, EEG, 3 references
[9] doi:10.1371/journal.pone.0358552 [code]
High-frequency stimulation-induced secondary hyperalgesia shifts temporal order judgment and increases precuneus nodal degree in healthy adults.
Journal: PloS one
In common: ICLabel, EEGLAB, FieldTrip, 2 other tools, EEG
[10] doi:10.1162/imag.a.1229 [code]
40 Hz audiovisual stimulation improves sustained attention and related brain oscillations.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: ICLabel, EEGLAB, FieldTrip, 2 other tools, EEG

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.