Pneumonia Detection Using Transfer Learning (CNN)

Emplement of Deep Learning (DL) models particularly Convolutional Neural Networks (CNNs) to detect pneumonia from Lung X-rays of patients. CNN-based VGG16 model is used to classify chest X-ray images of pneumonia patients and normal patients and predict patients have with pneumonia or not .

Datasets

with Kaggle dataset available from here.It consists of 5863 X-ray images of lungs taken on a group of paediatric patients . All of images have been labeled by 2 specialists to minimize labeling error risk. The data was already split for us into training, validation and test datasets.

Detection

In this project we'll Use Transfer learning using VGG16 model and customize our model to predict whether a person has Pneumonia.

• NORMAL IMAGE

• PNEUMONIA IMAGE

Then transfer learning was used with convolutional neural networks by utilising VGG16 as the base model for image classification. The model was able to achieve 94.5% accuracy on the validation set.

References

https://github.com/Sanket2311/Pneumonia-Classification-using-Transfer-learning