# DeepECG

This repository hosts the project for the Biomedical Signal and image Processing (BiSP) Lab at the University of Milan (Italy), focusing on ECG-based biometric recognition. The project is based on the methodology described by (Deep-ECG article) and utilizes the Smart Health for Assessing the Risk of Events via ECG Database.

## Requirements

To start the program you're gonna need:

* **Python 3.9** -> more info you can find https://www.python.org/downloads/release/python-390/ 

* **pip** -> more info about installation you can find https://pip.pypa.io/en/stable/cli/pip_install/ 

* **git** -> more info you can find https://git-scm.com/download/win

* **virtual environment (venv)** -> more info you can find https://virtualenv.pypa.io/en/stable/installation.html

## Seting up the virtual enviroment 

1. in Command Line go to cloned DeepECG repository. Create virtual environment - after this you shuld have aditional folfer **venv**

```commandline

C:\path\DeepECG> python -m venv C:\path\DeepECG\venv

```

2. activate environment - after this step *(venv)* should appear before path in the line

```commandline

C:\path\DeepECG> C:\path_to_program\DeepECG\venv\Scripts\activate

```

3. install requirements

```commandline

(venv) C:\path\DeepECG> pip install -r requirements.txt

```

## Pipeline

1. preprocessing and saving the data

```commandline

(venv) C:\path\DeepECG> python preprocess 'path/to/data' 'path/to/saving/directory'

```

2. running the experiments

```commandline

(venv) C:\path\DeepECG> python run_exp_1.py 'path/to/preprocessed/data'

```

## Authors

Zofia Mizgalewicz

Christian R. Cuenca

Dr. Massimo W. Rivolta

Prof. Roberto Sassi

Prof. Fabio Scotti

## Links

To see whats happaning in the [BiSP](https://bisp.di.unimi.it/)

For more details on the dataset, visit [SHAREEDB](https://physionet.org/content/shareedb/1.0.0/)