A Combined Approach Using Image Processing and Deep Learning to Detect Pneumonia from Chest X-Ray Image

Pneumonia is an epidemic disease that is needed to be detected in the early stage to prevent unfortunate deaths. Traditional methods take a lot of time to detect the disease. With the introduction of Medical Imaging, the detection of disease has been accelerated by using chest x-ray image. But it also requires the availability of an expert and experienced radiologist in order to interpret the x-ray image accurately. Sometimes, manual interpretation is affected by various kinds of artifact in X-ray or Optical Coherence Tomography(OCT) images. For this reason in our paper, we have proposed a combined approach using Image Processing and either VGG-16 or VGG-19, variants of Deep Convolutional Neural Network for automatic detection of pneumonia from Chest X-ray image. We have used Mendeley OCT and Chest X-Ray dataset to evaluate our model. We have achieved an accuracy of 96.2% using VGG-16 and 95.9% using VGG-19, both of which outperform transfer learned InceptionV3 benchmark model used on this dataset which has an accuracy of 92.8%.