marcoDige / Anomalous

Automatic analyzer of EGC signals built with Keras and HeartPy

A faster approach to ECG analysis in emergency situations

This is the official repository of the "A faster approach to ECG analysis in emergency situations" project for the 2020 "Progetto di Ingegneria Informatica" exam. The authors are:

• Luigi Fusco
• Ian Di Dio Lavore
• Marco Di Gennaro

Anomalous is a full stack ECG analysis system. It is capable of detecting tachycardia, bradycardia and several other heartbeat anomalies, more specifically left bundlebranch block beat, right bundle branch block beat, atrial premature beat, supraventricular premature orectopic beat, premature ventricular contraction and R-on-T premature ventricular contraction, with varying degree of precision.

Installation

Anomalous is build using the python programming language and makes use of several libraries and frameworks. It supports pipenv for the easy creation of ready to use virtual environments. pipenv requires python and pip to be installed in your sistem, and can be installed with:

python -m pip install pipenv

To download the project and setup and start the virtual environment type:

git clone https://github.com/luigifusco/Anomalous.git
cd Anomalous
pipenv install
pipenv shell

Alternatively all libraries can be installed in the current python environment with:

python -m pip install -r requirements.txt

A full version of the data folder is available here. Please replace the folder coming with this repo before running the demos.

To run a demo of the anomaly detection go in the Anomalous folder and type:

python main.py

To see she WIP GUI type

python gui_launcher.py

The project was build with and tested on python 3.7. At the time of writing tensorflow does not support python 3.8.

Special thanks

We would like to thank Eleonora D'Arnese and Marco D. Santambrogio for their continuous support to the project.

Disclaimer

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.