Diagnosis of COVID-19 from Multimodal Imaging Data Using Optimized Deep Learning Techniques

Abstract

COVID-19 had a global impact, claiming many lives and disrupting healthcare systems even in many developed countries. Various mutations of the severe acute respiratory syndrome coronavirus-2, continue to be an impediment to early detection of this disease, which is vital for social well-being. Deep learning paradigm has been widely applied to investigate multimodal medical image data such as chest X-rays and CT scan images to aid in early detection and decision making about disease containment and treatment. Any method for reliable and accurate screening of COVID-19 infection would be beneficial for rapid detection as well as reducing direct virus exposure in healthcare professionals. Convolutional neural networks (CNN) have previously proven to be quite successful in the classification of medical images. A CNN is used in this study to suggest a deep learning classification method for detecting COVID-19 from chest X-ray images and CT scans. Samples from the Kaggle repository were collected to analyse model performance. Deep learning-based CNN models such as VGG-19, ResNet-50, Inception v3 and Xception models are optimized and compared by evaluating their accuracy after pre-processing the data. Because X-ray is a less expensive process than CT scan, chest X-ray images are considered to have a significant impact on COVID-19 screening. According to this work, chest X-rays outperform CT scans in terms of detection accuracy. The fine-tuned VGG-19 model detected COVID-19 with high accuracy-up to 94.17% for chest X-rays and 93% for CT scans. This work thereby concludes that VGG-19 was found to be the best suited model to detect COVID-19 and chest X-rays yield better accuracy than CT scans for the model.

Comparative Analysis on Available Technique for the Detection of Covid-19 through CT-Scan and X-Ray using Machine Learning: A Systematic Review

“…They found only 485 clinical notes available for 535 images. Seven of them used only Kaggle dataset to collect their data for training or testing purpose of developed model [35,38,42,44,46,49,51]. Three of them used only GitHub repository which was developed by Dr. Joseph Cohen [33,43,48], other four of them used both of the Kaggle and GitHub database [34,47,50,54].…”

“…Data Source Data Volume [42] Kaggle 2482 CT scans image and 31 Covid positive along with 10,192 normal X-ray image [48] GitHub 108,948 images [49] Kaggle 3829 X-rays and 3829 X-rays [43] GitHub 535 CT and X-ray images and 485 clinical notes related with them [37] Mendeley Data 17,599 images [44] Kaggle 2481 images [46] Kaggle 2357 CT scan data, 2515 chest X-ray data and 2400 CT and chest X-ray hybrid data [34] Kaggle, GitHub 5856 chest X-ray & CT dataset [54] Kaggle, GitHub 2905 unique images for X-ray 617,775 images from 4154 patients [41] Mendeley Data 8,055 CT scan and 9,544 X-ray images [51] Kaggle 2591 mixed data [33] GitHub 17100 X−ray and CT images [47] Kaggle, GitHub 168 Covid and 168 normal cases for the both Xray and CT scan images [36] GitHub Kaggle and Git Hub 723 X-ray and 3228 CT scans images [38] Kaggle 400 chest X-ray images, and 400 CT scan images [39] Wonkwang University Hospital(WKUH) and Chonnam National University Hospital (CNUH), Italian Society of Medical and Interventional Radiology(SIRM) public database.…”

“…. For example deep learning based Convolutional Neural Network, Inception v3, VGG-19, ResNet-50 and Xception models are used by Mukhi S.E et al[42].After pre-processing the accuracy of models are analysed and estimated by obtained accuracy. They found that fine-tuned VGG-19 model has a highest accuracy at detection of covid-19.…”

Automatic detection of infected areas of CT images in COVID-19 patients using Inf-Seg-Net

“…A dataset of medical picture labels is used to train the model, and it is then tweaked to function as well as possible. Mukhi et al [34] Information from many imaging modalities, such as CT scans, X-rays, and molecular imaging, is included in the multimodal data. CNN models such the VGG-19, ResNet-50, Inception v3 and Xception models are utilised to extract pertinent features from the multimodal data and classify COVID-19 instances using optimised DL approaches.…”

Design of a Multimodal Detection System Tested on Tea Impurity Detection

“…Multimodal imaging systems find diverse applications across various fields due to their ability to provide comprehensive insights by combining different imaging techniques. Some of the key applications include medical diagnostics [1][2][3][4], biomedical research [5], environmental and earth sciences [6,7]. In addition, the detection content in the industry is becoming more and more complex, and the information obtained by a single sensor cannot meet the detection needs.…”