Deep-Learning-Assisted Highly-Accurate COVID-19 Diagnosis on Lung Computed Tomography Images

TL;DR

This paper presents a deep learning approach that enhances COVID-19 diagnosis accuracy from lung CT images by improving data quality and addressing class imbalance, achieving high performance on benchmark datasets.

Contribution

The authors introduce a novel data quality control pipeline using GANs and sliding windows, along with class-sensitive loss functions to improve COVID-19 detection accuracy.

Findings

Achieved over 0.983 MCC on benchmark datasets.

Effective handling of data quality issues with GAN-based pipeline.

Addressed dataset class imbalance with specialized loss functions.

Abstract

COVID-19 is a severe and acute viral disease that can cause symptoms consistent with pneumonia in which inflammation is caused in the alveolous regions of the lungs leading to a build-up of fluid and breathing difficulties. Thus, the diagnosis of COVID using CT scans has been effective in assisting with RT-PCR diagnosis and severity classifications. In this paper, we proposed a new data quality control pipeline to refine the quality of CT images based on GAN and sliding windows. Also, we use class-sensitive cost functions including Label Distribution Aware Loss(LDAM Loss) and Class-balanced(CB) Loss to solve the long-tail problem existing in datasets. Our model reaches more than 0.983 MCC in the benchmark test dataset.