Data Augmentation and CNN Classification For Automatic COVID-19 Diagnosis From CT-Scan Images On Small Dataset

TL;DR

This paper introduces a novel data augmentation technique using multiple Hounsfield Unit windows and an ensemble of CNN models to improve COVID-19 diagnosis accuracy from small CT scan datasets.

Contribution

It presents a unique data augmentation method and an ensemble approach with 2D CNNs to enhance classification performance on limited data.

Findings

Achieved 93.39% patient classification accuracy.

Effective data augmentation improves generalization on small datasets.

Ensemble of CNNs boosts diagnostic accuracy.

Abstract

We present an automatic COVID1-19 diagnosis framework from lung CT images. The focus is on signal processing and classification on small datasets with efforts putting into exploring data preparation and augmentation to improve the generalization capability of the 2D CNN classification models. We propose a unique and effective data augmentation method using multiple Hounsfield Unit (HU) normalization windows. In addition, the original slice image is cropped to exclude background, and a filter is applied to filter out closed-lung images. For the classification network, we choose to use 2D Densenet and Xception with the feature pyramid network (FPN). To further improve the classification accuracy, an ensemble of multiple CNN models and HU windows is used. On the training/validation dataset, we achieve a patient classification accuracy of 93.39%.