GACDN: generative adversarial feature completion and diagnosis network for COVID-19

Abstract

Background The outbreak of coronavirus disease 2019 (COVID-19) causes tens of million infection world-wide. Many machine learning methods have been proposed for the computer-aided diagnosis between COVID-19 and community-acquired pneumonia (CAP) from chest computed tomography (CT) images. Most of these methods utilized the location-specific handcrafted features based on the segmentation results to improve the diagnose performance. However, the prerequisite segmentation step is time-consuming and needs the intervention by lots of expert radiologists, which cannot be achieved in the areas with limited medical resources. Methods We propose a generative adversarial feature completion and diagnosis network (GACDN) that simultaneously generates handcrafted features by radiomic counterparts and makes accurate diagnoses based on both original and generated features. Specifically, we first calculate the radiomic features from the CT images. Then, in order to fast obtain the location-specific handcrafted features, we use the proposed GACDN to generate them by its corresponding radiomic features. Finally, we use both radiomic features and location-specific handcrafted features for COVID-19 diagnosis. Results For the performance of our generated location-specific handcrafted features, the results of four basic classifiers show that it has an average of 3.21% increase in diagnoses accuracy. Besides, the experimental results on COVID-19 dataset show that our proposed method achieved superior performance in COVID-19 vs. community acquired pneumonia (CAP) classification compared with the state-of-the-art methods. Conclusions The proposed method significantly improves the diagnoses accuracy of COVID-19 vs. CAP in the condition of incomplete location-specific handcrafted features. Besides, it is also applicable in some regions lacking of expert radiologists and high-performance computing resources.

A Review Study of the Deep Learning Techniques used for the Classification of Chest Radiological Images for COVID-19 Diagnosis

“…GANs were widely used by researchers in different domains like data augmentation, image segmentation and so on. Zhu et al ( Zhu et al, 2021 ) applied an optimization GANs methods (GACDN) to improve the feature extraction step. He et al ( He et al, 2021 ) developed an evolvable GAN framework by utilizing three different mutation operators, weight clipping and JS divergence, the Wasserstein distance, and so on, for the target of overcoming the limitation and problems of GANs.…”

“…Zhu et al ( Zhu et al, 2021 ) proposed a method based on GANs that generates handcrafted features by radiomic counterparts of CT images, called generative adversarial feature completion and diagnosis network (GACDN). Their framework focuses on the problem of single-view diagnostic frameworks ignoring basic information in handcrafted features for a particular location.…”

A review of medical image-based diagnosis of COVID-19

“…However, machine learning, especially deep learning, is primarily dependent on a large number of images. Therefore, many researchers used weakly supervised learning methods for COVID-19 detection and diagnosis tasks [13,14]. Weakly supervised learning is intermediate between supervised and unsupervised learning, and it can obtain favorable results based on a small amount of data.…”

Explorando Características Radiômicas Profundas em Modelos de Aprendizado Profundo por meio de Imagens Médicas de COVID-19

“…Em Zhu et al (2021) é abordado o trabalho GACDN: generative adversarial feature completion and diagnosis network for COVID-19, que apresenta um framework para multivisualização incompleta das características artesanais geradas para diagnosticar COVID-19. Zhu et al (2021) propõem o uso das Redes Adversariais Generativas (do inglês, Generative Adversarial Network -GAN) com três redes embutidas, a geradora, discriminadora e a rede consistente com a doença. A rede geradora sintetiza o local específico da característica e a rede discriminadora avalia a qualidade da característica.…”