Experiment Data Extraction – MATLAB UI

A MATLAB-based graphical user interface (GUI) for extracting, organizing, and processing multimodal experimental data from .xdf files across multiple subjects, activities, and experimental conditions.

This tool was developed to support large-scale human performance and sensorimotor experiments involving synchronized physiological, biomechanical, and behavioral data streams.

---

Author

Jeevan Jayasuriya
NeuroErgonomics Lab
University of Wisconsin–Madison

---

Overview

This repository provides a unified MATLAB UI for:

• Selecting subject ranges (e.g., S1–S10)
• Selecting experimental activities (single activity or all activities)
• Selecting data streams (e.g., fNIRS, EMG, Force Plate, Eye Tracking, Xsens kinematics)
• Extracting synchronized time series and time stamps from .xdf files
• Automatically routing extracted data to dedicated processing functions

The UI is designed to minimize manual file handling and enforce consistent data organization across large experimental datasets.

---

Key Features

• Interactive MATLAB UI built with uifigure
• Batch extraction across:
  • Multiple subjects
  • Multiple activities
  • Pre/Post sessions
  • Stimulated / Non-stimulated trials
• Modular processing architecture
• Supports multimodal data streams commonly used in neuroergonomics research
• Easily extendable to new streams or processing functions

---

Repository Structure

UI and Core Logic

• Main_UI_V1.m
  Main UI entry point (Data_extraction_functions_v12) and data extraction logic.

• load_xdf.m
  XDF file loader used throughout the pipeline.

Processing Modules

• processfNIRS_V1.m
• processHomer3_V1.m
• processEMG_V1.m
• processHeartRate_V1.m
• processForcePlate_V1.m
• processEyeTracking_V1.m
• processLinearSegmentKinematics_V1.m
• processAngularKinematics1_V1.m
• processCenterOfMass1_V1.m
• processEulerDatagram1_V1.m
• processTimeFrames_V1.m
• processVideoFrames_V1.m
• processQuaternion.m

Utilities

• Extract_ECG_from_XDFandActiheart.m
• CombineMATFiles.m

---

Supported Data Streams

• Brain data: fNIRS, Homer3
• Muscle & physiology: EMG, Heart Rate (including HRV metrics)
• Biomechanics: Force Plate, Center of Mass
• Motion capture (Xsens): Linear and Angular kinematics, Euler angles, Quaternions
• Perceptual & auxiliary: Eye Tracking, Video frames, Timestamp streams

Channel and segment options are dynamically updated in the UI based on the selected data type.

---

How to Run

1. Clone or download the repository
2. Add the repository to your MATLAB path:

   addpath(genpath(pwd))

3. Launch the UI:

   Main_UI_V1

4. Use the UI to select subjects, activities, and data types, then click Extract Data.

---

Outputs

Depending on the selected data type, processing functions may export processed data to MATLAB variables, save analysis-ready files, or generate plots. Output behavior is defined within each process*.m file.

---

Extending the Tool

To add a new data stream:

1. Add the stream name to the UI dropdown
2. Define channel/segment options
3. Implement a new processYourStream_V1.m
4. Register it in the switch dataType block

---

Notes

• File paths and storage locations are intentionally not hard-coded
• Users are expected to configure their own data directories externally
• Experimental data are not included in this repository

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

Acknowledgments

Developed as part of ongoing research at the NeuroErgonomics Lab, University of Wisconsin–Madison.