Skip to content

Latest commit

 

History

History
75 lines (51 loc) · 3.14 KB

README.md

File metadata and controls

75 lines (51 loc) · 3.14 KB
Raw

COPERIA: AI Models for Voice Signal Analysis in PASC Patients

About

This repository provides an implementation for the data pipelines and AI models used in the COPERIA project.

The COPERIA project aims to develop and clinically validate a comprehensive multidisciplinary platform that utilizes artificial intelligence for the diagnosis, empowerment, and clinical management of Post-Acute Sequelae of SARS-CoV-2 (PASC) patients. The clinical study conducted in the context of the COPERIA project received ethical approval from the Clinical Research Ethics Committee of Galicia, and all procedures were conducted in compliance with the ethical principles outlined in the Declaration of Helsinki. Informed consent was obtained from all participants prior to their involvement in the study. The study was registered in the US Clinical Trials Registry under the code [NCT05629793].

The project is developed by the Multimedia Technologies Group at the atlanTTic Research Center, Universidade de Vigo, in collaboration with the "Persistent COVID Unit of the Ourense Hospital" and primary care centers in the health area.

Patient distribution of COPERIA project.

Fig. 1: Patient distribution of the clinical trial

Recordings distribution of COPERIA project.

Fig. 2: Recordings distribution of the clinical trial

See more details here: COPERIA-Dashboard

Features

  • Data pipelines for processing voice and metadata data from the COPERIA project.
  • AI models for voice signal analysis in PASC patients.
  • Data visualization tools for exploring the data.
  • Data analysis tools for extracting insights from the data.

Getting Started

Recommended Python version: 3.9

  1. Clone the repository:
git clone https://github.com/JMasr/corilga_api.git
  1. Navigate to the project directory:
cd corilga_api
  1. Import the envairoment using conda:
conda env create -f env.yml
  1. Install requirements:
pip install -r requirements.txt
  1. Use exp_config.json to set the paths to the data and models.

Acknowledgements

  • This work was supported by the CONECTA COVID programme, co-financed by the European Regional Development Fund (ERDF) within the Galicia ERDF operational programme 2014-2020 as part of the EU’s response to the COVID19 pandemic, and Axencia Galega de Innovacion (GAIN).

  • This work has received financial support from the Xunta de Galicia (Centro singular de investigación de Galicia accreditation 2019-2022).

  • This research has been funded by the Galician Regional Government under project ED431B 2021/24"GPC".

  • Thanks to the “Unidad de COVID Persistente del Hospital de Ourense” and the patients involved in the study.