Classification of Covid and Pneumonia using Deep Learning

Project Title

Classification of Covid and Pneumonia using Deep Learning

Project Area of Specialization Artificial IntelligenceProject Summary

The Project is a model that is based on Deep Learning which is trained on some data sets to provide an accuracy above  95%, we have created 3 scenarios that take some dataset on some  ratios  and provide the accuracy. The result is shown in binary i.e 0 or 1 for Covid-19 or Normal.  We have also presented some charts for graphical representation at the end of the model.

Project Objectives

The objective is for the practicing  doctors so that they can identify which chest x-ray is of Covid-19 or Normal. Since the pandemic started many patients were admitted by mistakes of the practicing doctors so to avoid that we created the model which is trained on the datasets to identify for them.

Project Implementation Method

Implementation method is simple why ? because we wanted to make the model in a simple manner so even a non technical person can understand what is happening in it the steps are as follows :

1- Created senerios on Jupyter Notebook.

2- Uploaded them on dropbox so we can easily use them in our Colabs notebook.

3- Imported the libraries we need for the model in colabs.

4- Trained the model in google colabs so that even if we need the extra resources we could buy them.

Benefits of the Project

Benefits :

1- Easy to use for the practicing docters since they just have to use the x-ray name.

2- provides an accuracy above  95% in medical field thats appropriate for a model.

3- we made it in such a way that it can be more advanced by adding more layers and images to the dataset

4- workload would be reduced in testing the patients.

5- the model is not just for Covid-19 it can be used in other diseases too.

Technical Details of Final Deliverable

There is no such rocket science in the technical details of our project since we wanted to make it simple in the pandemic for all users . 

How its working -> 

Simply the datasets are being called from the dropbox and imported in the notebooks.

Then the model trains and creates 2 class indices 0 & 1.

Then we can provide pictures to it and get results based on binary numbers.

Final Deliverable of the Project Software SystemCore Industry MedicalOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	1817
google colab pro	Miscellaneous	1	1817	1817