Covid Detectioin via X-ray Image Processing

Project: Developed a deep learning system for COVID-19 detection via X-ray using CNN

1. Supervisor: Mam, Meher-Un-Nisa

2. Group Members: Saad Ali Zafar Sajid

3. Covid Detection Via X-ray Image Processing

4. Covid is spreading rapidly throughout the world, best standard methods for detecting corona is PCR test, however, the facilities for this are limited . By using Deep Learning Model Inception V3,a system is developed to give a self detection method to save crucial resources. . Introduction

5. To provide tool for Covid detection based on X-ray Image Processing. Collection and analysis of data set. Purpose and train deep learning algorithm for Covid detection. Performance evaluation of proposed model. Project Goal 1

6. METHODOLOGY

7. ‹#› METHODOLOGY Data sets were obtained from Kegel . Feature extracting algorithm INCEPTION V3 , a Deep Learning Algorithm is used. Extracted features/Labels will be feed to a SVM model Classification Model, SVM(Support Vector Machine) will save all the labels of training images, will provide offline service instead of retraining the system every time. A GUI is build that helps user to upload respective image. In that very same GUI, result box displays the predicted result.

8. X-ray Image of Covid & Normal Patient Above are the X-rays for pneumonia and Covid positive patient. For a normal eye it seems very similar to rest of the chest infection diseases. A proper expert body is required to distinguish them for the Covid positive X-ray.

9. IMPLEMENTATION We have used MATLAB2021B GUI App Designer for the GUI

10. FINAL DESIGN GUI

11. TRAINING OF DATASET

12. TESTING OF DATASET

13. DEEP LEARNING MODEL

14. INCEPTION V3 The Inception V3 is a deep learning model based on Convolutional Neural Networks, which is used for image classification. The inception V3 is a superior version of the basic model Inception V1 which was introduced as GoogLeNet in 2014. As the name suggests it was developed by a team at Google. I nception V 3 is a Convolution Neural Network for assisting in image analysis and object detection. It extracted features from all images using filter, pooling , convolution to get maximum labels to be sent to SVM classifier.

15. SVM Support vector machines (SVMs) are a set of supervised learning methods used for classification, regression and outliers detection. The advantages of support vector machines are: Effective in high dimensional spaces. Still effective in cases where number of dimensions is greater than the number of samples. The objective of the Support vector machine (SVMs) algorithm is to find the number of features that distinctly classifies the data points. It is used to retrieve trained data set in an offline mode in order to save computational time.

16. CONFUSION MATRIX A confusion matrix is a technique for summarizing the performance of a classification algorithm.

17. ACCURACY Accuracy with the help of confusion matrices represent counts from predicted and actual values. The output “ TN ” stands for True Negative which shows the number of negative examples classified accurately. Similarly, “ TP ” stands for True Positive which indicates the number of positive examples classified accurately. The term “ FP ” shows False Positive value, i.e., the number of actual negative examples classified as positive; and “ FN ” means a False Negative value which is the number of actual positive examples classified as negative. One of the most commonly used metrics while performing classification is accuracy. The accuracy of a model (through a confusion matrix) is calculated using the given formula below. ACCURACY = ( TP + TN ) (TP+TN+FP+FN) 102 + 134 = 236 = 0.96 X 100= 96.3% 102+3+6+4 245

18. Thank you