SSNR ADL Classification

This repo includes anonymized labeled data for activities of daily living in spinal cord injury individuals. There are also introductions to classification pipelines.

Cloning the Repository

To get started with this project, first, you need to clone the repository to your local machine. To do so, first install git on your machine: https://git-scm.com/downloads

Then open your terminal and run the following command:

git clone https://github.com/SCAI-Lab/ssnr_adl_classification.git

Setting Up the Virtual Environment

After cloning the repository, navigate into the project directory:

cd <directory to the cloned repo>/ssnr_adl_classification

It's highly recommended to use a virtual environment to manage your project dependencies. Follow these steps to create and activate a virtual environment with Python 3.10.13:

1. Create a virtual environment:

   python3.10 -m venv venv

2. Activate the virtual environment:

   • On macOS/Linux:

     source venv/bin/activate

   • On Windows:

     .\venv\Scripts\activate

Installing Required Libraries

Once the virtual environment is activated, install the necessary libraries using pip and the requirements.txt file:

pip install -r requirements.txt

This will ensure that all the dependencies for the project are installed and properly configured.

---

Associating Virtual Environment with Jupyter Kernel

To use the created virtual environment with Jupyter notebooks, you need to associate it with a Jupyter kernel. Follow these steps:

1. Install the IPython kernel for the virtual environment:

   python -m ipykernel install --user --name=venv

2. Start Jupyter Notebook:

   jupyter notebook

3. Select the Kernel in Jupyter Notebook:

   • Open the notebook you want to work on.
   • Go to the Kernel menu, select Change kernel, and choose venv.

---

Make sure to deactivate the virtual environment when you're done working on the project:

deactivate

Instructions and tutorials

In the notebooks folder, you will find the following Jupyter notebooks with instructions for loading data:

• tutorial_signal_processing.ipynb: Instructions for signal processing.
• tutorial_classification.ipynb: Instructions for creating a classification pipeline.
• tutorial_lime.ipynb: Instructions for Explaining a model using LIME analysis.
• tutorial_shap_exercise.ipynb: Instructions for explaining a model with SHAP values
• tutorial_shap_solution.ipynb: Solution of the exercise

Shapley Analysis Exercise

Note: The ecg data can be requested from SCAI lab.

Credits: The ECG encoder and decoder were adapted from the paper: Kuznetsov, V. V., et al. "Interpretable feature generation in ECG using a variational autoencoder." Frontiers in genetics 12 (2021): 638191.