A Deep Learning Model with Self-Supervised Learning and Attention Mechanism for COVID-19 Diagnosis Using Chest X-ray Images

Abstract

The SARS-CoV-2 virus has spread worldwide, and the World Health Organization has declared COVID-19 pandemic, proclaiming that the entire world must overcome it together. The chest X-ray and computed tomography datasets of individuals with COVID-19 remain limited, which can cause lower performance of deep learning model. In this study, we developed a model for the diagnosis of COVID-19 by solving the classification problem using a self-supervised learning technique with a convolution attention module. Self-supervised learning using a U-shaped convolutional neural network model combined with a convolution block attention module (CBAM) using over 100,000 chest X-Ray images with structure similarity (SSIM) index captures image representations extremely well. The system we proposed consists of fine-tuning the weights of the encoder after a self-supervised learning pretext task, interpreting the chest X-ray representation in the encoder using convolutional layers, and diagnosing the chest X-ray image as the classification model. Additionally, considering the CBAM further improves the averaged accuracy of 98.6%, thereby outperforming the baseline model (97.8%) by 0.8%. The proposed model classifies the three classes of normal, pneumonia, and COVID-19 extremely accurately, along with other metrics such as specificity and sensitivity that are similar to accuracy. The average area under the curve (AUC) is 0.994 in the COVID-19 class, indicating that our proposed model exhibits outstanding classification performance.

Self-supervised learning for medical image classification: a systematic review and implementation guidelines

“…Jung et al created a functional connectivity matrix between pairs of region-of-interest in rs-fMRI for each subject, and created a masked auto-encoder task by randomly masking out different rows and columns of the matrix for restoration 66 . Two of the five studies compared their approach to models without self-supervised pre-training and reported slight relative improvements in performance of 1.12% 67 and 0.690% 63 .…”

“…The vast majority of the studies explored the clinical domain of radiology (47/79), of which 27 were focused on investigating abnormalities on chest imaging such as pneumonia, COVID-19, pleural effusion and pulmonary embolism (see Table 1 ). The choice of this domain is likely a combination of the availability of large-scale public chest datasets such as CheXpert 73 , RSPECT 74 , RadFusion 75 and MIMIC-CXR 76 , as well as the motivation to solve acute or emerging healthcare threats, which was the case during the coronavirus pandemic 45 , 46 , 48 , 67 , 77 – 83 . The second most prevalent clinical domain was pathology (12/79).…”

Automatic feature extraction by supervised and contrastive self-supervised learning based on wavelet and hard negatives to detect HIFU lesion area

“…[4][5][6][7][8][9][10][11][12][13] However, medical image processing faces unique challenges, notably limited data and labeling; consequently, recent efforts have aimed to reduce reliance on data annotation across various medical image types [14][15][16][17]. Studies have predominantly focused on self-supervised methods for X-ray images [18][19][20][21][22][23] with comparatively fewer articles addressing MRI [24][25][26][27] and CT-scan [28][29][30][31][32], and notably fewer on ultrasound images. [33,34] Articles discussing HIFU control and monitoring include [35] presenting a method utilizing ultrasound signals as input to a feedforward neural network for lesion area detection.…”

Preliminary Stages for COVID-19 Detection Using Image Processing

“…While most researchers use two or more techniques, others use only one technique. For instance, in [ 67 ], flipping, zooming, and width shifting were used as data augmentation techniques to reduce the bias caused by the properties of CXR images. Joshi et al [ 66 ] used image scaling and rotation to increase the original dataset CXR images by five.…”