Autonomic Disease Diagnostics System for Chest Radiographs

Project Title

Autonomic Disease Diagnostics System for Chest Radiographs

Project Area of Specialization Artificial IntelligenceProject Summary

The hospital / health care’s need’s some intelligent system for diagnosing the disease from the x rays. If we have millions of x rays of people’s and we check the disease manually, it will consume so much time and energy. So, we have a problem solution that we can create a smart system that can diagnose the disease. The machine can take x ray image as an input and then diagnose what’s the disease is, like: It’s Covid, Pneumonia, TB etc. This system can have used the technology of Deep Learning (DL) Model. We have disease classification model and we train the model with the help of x ray images. This system consists of Web Application for the hospital usage in which user can only browse the x ray image, the backend part should be cover with the Deep Learning (DL) Model.

Project Objectives

We have an Objective that:

1. To design a Web Application for the hospitals that can deal this pandemic or diseases patients every day.
2. To achieve the application which can create an effective system that can help the doctors to diagnose the disease more efficiently.
3. To the creation and proper implementation of this system development, the doctors shall be facilitated immensely and more patients can be diagnosed quickly resulting in efficient treatment in no time and delays in diagnosing and treatment shall be reduced to greater extent.
4. To conclude the final result that appear, when the application is processed successfully. We feel satisfied in a manner that eventually we can also help the hospitals or the patients in the creation and obtaining results through thus system than can further provide guidance and help the doctors to diagnose the disease in quick time.

Project Implementation Method

METHODOLOGY / IMPLEMENTATION OF PRODUCT:

Data Retrieval:

In the first phase, we can collect the data from the external websites they can provide the x ray images. Also we collect the data from the hospitals. We have too many files of the x ray images from the different sources.

Data Augmentation:

In this phase, if the collection of data is not enough for our model so, we can use the technique of augmentation in which we have one image and we create multiple images from one image like rotating the image, scaling the image and also duplicates from the four sides of the image because with this rotation process we have all four sided angle of the image that we will give to the Neural Network for the training that’s why we use this technique of augmentation. We have another advantage is that when we do this augmentation of data we increase the dataset from the limited data to more enough for our Model.

Testing a Model:

In this phase, after the completion of data cleaning process now we have the formatted data in our one file. Now, we created a CNN based Deep Learning Model in which we have two parts training and testing part we give the more data to training part and some data to testing part. After train our model we check the accuracy of our model. When if the model accuracy not good enough we use the different techniques to make our model efficient.

Web Application of Product:

In the final phase, we create the web based application for users in which we have the multiple pages that describes the detailed information about the disease and the symptoms of the disease. We have one practical page which contains the browse the image feature that user can select the chest x ray image from the computer and hit the button below our Deep Learning Model will execute from behind and they predict the disease with respect to the symptoms that the person has.

Benefits of the Project

In the pandemic we have millions of patient’s chest x rays in the hospital. The problem raised that the doctors have to check the x rays manually to diagnose the disease. It’s will take too much time. All the hospital’s need’s some smart systems to check the x rays digitally. The important of this problem must increase because of the diseases. The Covid, Pneumonia and TB all have the chest x rays so, the doctors check them manually what’s the actual disease is that’s why they need’s some smart system that can predict what the actual disease is with the help of x ray images.

Technical Details of Final Deliverable

Phase I – V:

We use Jupyter Notebook as an editor.

Libraries for Phase I – V:

Numpy, Pandas, Matplotlib, Seaborn, Image, Sklearn, Preprocessing, model selection, tensorflow, keras, optimizers, OpenCV, Heroku etc.

Phase VI:

React JS, API, Firebase, Authentication etc.

Final Deliverable of the Project Software SystemCore Industry ITOther Industries Medical Core Technology Artificial Intelligence(AI)Other Technologies OthersSustainable 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)	46050
License Software's, API'S, Hosting Services	Equipment	5	9210	46050