Amrit

Amrit

MSCS, Georgia tech/ ML specialisation / Physician

COVCT: GSOC proposal for android application (CovCT) for analyzing COVID and non-COVID data.

1 minute read

Project proposal for Google GSOC in association with Emory university school of medicine.

Github page - https://github.com/Amritpal-001/CovCT

This repository has been created for the development of an android application (CovCT) for analyzing COVID and non-COVID data.

Mentors: Monjoy Saha

Brief Demo - COV_CT

Video demo link-

Demo Video

Overview: Coronavirus disease 2019 (COVID-19) is a highly contagious respiratory disease caused by SARS-CoV-2. Uncounted people around the world got affected due to this virus. Various articles have reported that COVID-19 and Pneumonia cases have very few distinct features. We aim to develop an AI-based Android Application, which will distinguish COVID-19 and Pneumonia cases using computed tomography (CT) images.

Present Status of the work: A classification architecture using TensorFlow has been developed using Python. It works on Linux, Mac, and Windows systems.

Deliverables: The student will work on the development of an AI-based android application. The students may explore the idea of federated learning techniques on Android phones.

Source Code: https://github.com/monjoybme/CovCT

Data

Data is available at the below links. You may need to register before downloading the data.

https://covid-segmentation.grand-challenge.org/COVID-19-20/

https://mosmed.ai/en/

https://wiki.cancerimagingarchive.net/display/Public/CT+Images+in+COVID-19

https://github.com/mr7495/COVID-CTset

https://www.kaggle.com/azaemon/preprocessed-ct-scans-for-covid19

The above links are mostly for COVID data. There might be some non-COVID cases. Please read the data description before using them.

Non-COVID data is available at the below link: https://wiki.cancerimagingarchive.net/display/NLST/National+Lung+Screening+Trial

You can also search on https://datasetsearch.research.google.com/ for more data.

You May Also Enjoy