Uncertainty-Aware Semi-supervised Method using Large Unlabelled and Limited Labeled COVID-19 Data

02/12/2021
by   Roohallah Alizadehsani, et al.
0

The new coronavirus has caused more than 1 million deaths and continues to spread rapidly. This virus targets the lungs, causing respiratory distress which can be mild or severe. The X-ray or computed tomography (CT) images of lungs can reveal whether the patient is infected with COVID-19 or not. Many researchers are trying to improve COVID-19 detection using artificial intelligence. In this paper, relying on Generative Adversarial Networks (GAN), we propose a Semi-supervised Classification using Limited Labelled Data (SCLLD) for automated COVID-19 detection. Our motivation is to develop learning method which can cope with scenarios that preparing labelled data is time consuming or expensive. We further improved the detection accuracy of the proposed method by applying Sobel edge detection. The GAN discriminator output is a probability value which is used for classification in this work. The proposed system is trained using 10,000 CT scans collected from Omid hospital. Also, we validate our system using the public dataset. The proposed method is compared with other state of the art supervised methods such as Gaussian processes. To the best of our knowledge, this is the first time a COVID-19 semi-supervised detection method is presented. Our method is capable of learning from a mixture of limited labelled and unlabelled data where supervised learners fail due to lack of sufficient amount of labelled data. Our semi-supervised training method significantly outperforms the supervised training of Convolutional Neural Network (CNN) in case labelled training data is scarce. Our method has achieved an accuracy of 99.60 CNN (trained supervised) has achieved an accuracy of 69.87 94

READ FULL TEXT

page 7

page 8

page 15

research
01/18/2021

Covid-19 classification with deep neural network and belief functions

Computed tomography (CT) image provides useful information for radiologi...
research
02/08/2017

An Adversarial Regularisation for Semi-Supervised Training of Structured Output Neural Networks

We propose a method for semi-supervised training of structured-output ne...
research
11/06/2022

Learning to Annotate Part Segmentation with Gradient Matching

The success of state-of-the-art deep neural networks heavily relies on t...
research
02/11/2022

Explainable COVID-19 Infections Identification and Delineation Using Calibrated Pseudo Labels

The upheaval brought by the arrival of the COVID-19 pandemic has continu...
research
02/04/2022

Video Violence Recognition and Localization using a Semi-Supervised Hard-Attention Model

Empowering automated violence monitoring and surveillance systems amid t...
research
05/24/2019

Semi-supervised GAN for Classification of Multispectral Imagery Acquired by UAVs

Unmanned aerial vehicles (UAV) are used in precision agriculture (PA) to...
research
02/27/2019

Zoho at SemEval-2019 Task 9: Semi-supervised Domain Adaptation using Tri-training for Suggestion Mining

This paper describes our submission for the SemEval-2019 Suggestion Mini...

Please sign up or login with your details

Forgot password? Click here to reset