Skip to content

The 2019 novel coronavirus (COVID-19) presents several unique features. While the diagnosis is confirmed using polymerase chain reaction (PCR), infected patients with pneumonia may present on chest X-ray and computed tomography (CT) images with a pattern that is only moderately characteristic for the human eye. Bilateral multiple lobular and sub…

License

Notifications You must be signed in to change notification settings

SUMEETRM/covid19-ai

Repository files navigation

The model provided here is to be used for research and educational purposes only. Do not use the model for self diagnosis

The 2019 novel coronavirus (COVID-19) presents several unique features. While the diagnosis is confirmed using polymerase chain reaction (PCR), infected patients with pneumonia may present on chest X-ray and computed tomography (CT) images with a pattern that is only moderately characteristic for the human eye. Bilateral multiple lobular and subsegmental areas of consolidation can be observed in COVID-19 patients. The following model aims to present a neural network aimed to detect COVID-19 cases through chest X-Rays. While a neural network to detect COVID-19 cases has been published, the following model has a much higher accuracy. Out model uses a sample of 100 COVID-19 positive cases and 100 COVID-19 negative cases, and has an accuracy of 91% . Of the true positive patients, the model had an accuracy of 100%, and of the true negative patients, the model had an accuracy of 80%. Given the fact that the data is limited, the model can be improved in the near future for rapid diagnosis of COVID-19 cases with an extremely low rate of false positives. While the model may still be in an initial phase of development, it can be put to practical use soon with enough training and accuracy. corona_deeplearning

If you would like to contribute to this project, you can email us at [email protected] or [email protected]

After a number of downloads, we are not making the code available on this repo to prevent any misuse. Please email us if you're interested.

Dataset

For covid19-ai, we have used the following repo for our data set. The images are being updated constantly https://github.com/ieee8023/covid-chestxray-dataset

Press

Times of India: https://timesofindia.indiatimes.com/city/mumbai/class-xi-students-work-on-model-to-spot-x-rays-of-covid-19-patients/articleshow/75151793.cms Your Story: https://yourstory.com/2020/04/coronavirus-class-11-students-machine-learning-model Republic World: https://www.republicworld.com/india-news/city-news/dias-class-xi-students-develop-model-to-test-covid-through-x-ray-scans.html

About

The 2019 novel coronavirus (COVID-19) presents several unique features. While the diagnosis is confirmed using polymerase chain reaction (PCR), infected patients with pneumonia may present on chest X-ray and computed tomography (CT) images with a pattern that is only moderately characteristic for the human eye. Bilateral multiple lobular and sub…

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages