COVID-19_detector

The project's goal is to use a Convolutional Neural Network (CNN) classifier to classify positive and negative COVID-19 cases from lung X-rays, and then optimise the model using gradient descent. The project entails training the CNN model on a large dataset of COVID-19 positive and negative lung X-rays in order to accurately classify new X-rays. To minimise the cost function and improve the model's accuracy, the gradient descent optimization technique is used. The trained model can be used to quickly and accurately diagnose COVID-19 using X-ray images, especially in areas where PCR tests are unavailable. The project has the potential to benefit the healthcare industry by assisting in the early detection and management of COVID-19 cases.

Current Accuracy ~ 93%

References:

1. Pre-Processed Data; used in Main Model
2. Normal Cases Image Dataset
3. COVID Positve Case Image Dataset
4. Prateek Narang

Technology:

1. Google Colab
2. Keras

Packages:

1. Matplotlib
2. Numpy
3. Pandas

Concepts:

1. CNN Classifier
2. Gradient Descent Optimizer
3. Training & Validation