Classification of COVID-19 patients from chest CT images using multi-objective differential evolution–based convolutional neural networks

European Journal of Clinical Microbiology & Infectious Diseases, Journal Year: 2020, Volume and Issue: 39(7), P. 1379 - 1389

Published: April 27, 2020

Early classification of 2019 novel coronavirus disease (COVID-19) is essential for cure and control. Compared with reverse-transcription polymerase chain reaction (RT-PCR), chest computed tomography (CT) imaging may be a significantly more trustworthy, useful, rapid technique to classify evaluate COVID-19, specifically in the epidemic region. Almost all hospitals have CT machines; therefore, images can utilized early COVID-19 patients. However, CT-based involves radiology expert considerable time, which valuable when infection growing at rate. Therefore, an automated analysis desirable save medical professionals' precious time. In this paper, convolutional neural networks (CNN) used COVID-19-infected patients as infected (+ve) or not (-ve). Additionally, initial parameters CNN are tuned using multi-objective differential evolution (MODE). Extensive experiments performed by considering proposed competitive machine learning techniques on images. shows that model good accuracy

Language: Английский

---

Application of deep learning for fast detection of COVID-19 in X-Rays using nCOVnet

Chaos Solitons & Fractals, Journal Year: 2020, Volume and Issue: 138, P. 109944 - 109944

Published: May 27, 2020

Language: Английский

---

AI4COVID-19: AI enabled preliminary diagnosis for COVID-19 from cough samples via an app

Informatics in Medicine Unlocked, Journal Year: 2020, Volume and Issue: 20, P. 100378 - 100378

Published: Jan. 1, 2020

Background: The inability to test at scale has become humanity's Achille's heel in the ongoing war against COVID-19 pandemic. A scalable screening tool would be a game changer. Building on prior work cough-based diagnosis of respiratory diseases, we propose, develop and an Artificial Intelligence (AI)-powered solution for infection that is deployable via smartphone app. app, named AI4COVID-19 records sends three 3-second cough sounds AI engine running cloud, returns result within two minutes. Methods: Cough symptom over thirty non-COVID-19 related medical conditions. This makes by alone extremely challenging multidisciplinary problem. We address this problem investigating distinctness pathomorphological alterations system induced when compared other infections. To overcome training data shortage exploit transfer learning. reduce misdiagnosis risk stemming from complex dimensionality problem, leverage multi-pronged mediator centered risk-averse architecture. Results: Results show can distinguish among coughs several types coughs. accuracy promising enough encourage large-scale collection labeled gauge generalization capability AI4COVID-19. not clinical grade testing tool. Instead, it offers anytime, anywhere, anyone. It also decision assistance used channel clinical-testing treatment those who need most, thereby saving more lives.

Language: Английский

---

Explainable Deep Learning for Pulmonary Disease and Coronavirus COVID-19 Detection from X-rays

Computer Methods and Programs in Biomedicine, Journal Year: 2020, Volume and Issue: 196, P. 105608 - 105608

Published: June 21, 2020

Language: Английский

---

Classification of the COVID-19 infected patients using DenseNet201 based deep transfer learning

Journal of Biomolecular Structure and Dynamics, Journal Year: 2020, Volume and Issue: 39(15), P. 5682 - 5689

Published: July 3, 2020

Deep learning models are widely used in the automatic analysis of radiological images. These techniques can train weights networks on large datasets as well fine tuning pre-trained small datasets. Due to COVID-19 dataset available, neural be for diagnosis coronavirus. However, these applied chest CT image is very limited till now. Hence, main aim this paper use deep architectures an automated tool detection and CT. A DenseNet201 based transfer (DTL) proposed classify patients COVID infected or not i.e. (+) (−). The model utilized extract features by using its own learned ImageNet along with a convolutional structure. Extensive experiments performed evaluate performance propose DTL scan Comparative analyses reveal that classification outperforms competitive approaches.

Language: Английский

---

Deep Learning for Classification and Localization of COVID-19 Markers in Point-of-Care Lung Ultrasound

IEEE Transactions on Medical Imaging, Journal Year: 2020, Volume and Issue: 39(8), P. 2676 - 2687

Published: May 13, 2020

Deep learning (DL) has proved successful in medical imaging and, the wake of recent COVID-19 pandemic, some works have started to investigate DL-based solutions for assisted diagnosis lung diseases. While existing focus on CT scans, this paper studies application DL techniques analysis ultrasonography (LUS) images. Specifically, we present a novel fully-annotated dataset LUS images collected from several Italian hospitals, with labels indicating degree disease severity at frame-level, video-level, and pixel-level (segmentation masks). Leveraging these data, introduce deep models that address relevant tasks automatic In particular, network, derived Spatial Transformer Networks, which simultaneously predicts score associated input frame provides localization pathological artefacts weakly-supervised way. Furthermore, new method based uninorms effective aggregation video-level. Finally, benchmark state art estimating segmentations biomarkers. Experiments proposed demonstrate satisfactory results all considered tasks, paving way future research data.

Language: Английский

---

Artificial Intelligence and COVID-19: Deep Learning Approaches for Diagnosis and Treatment

IEEE Access, Journal Year: 2020, Volume and Issue: 8, P. 109581 - 109595

Published: Jan. 1, 2020

COVID-19 outbreak has put the whole world in an unprecedented difficult situation bringing life around to a frightening halt and claiming thousands of lives. Due COVID-19's spread 212 countries territories increasing numbers infected cases death tolls mounting 5,212,172 334,915 (as May 22 2020), it remains real threat public health system. This paper renders response combat virus through Artificial Intelligence (AI). Some Deep Learning (DL) methods have been illustrated reach this goal, including Generative Adversarial Networks (GANs), Extreme Machine (ELM), Long/Short Term Memory (LSTM). It delineates integrated bioinformatics approach which different aspects information from continuum structured unstructured data sources are together form user-friendly platforms for physicians researchers. The main advantage these AI-based is accelerate process diagnosis treatment disease. most recent related publications medical reports were investigated with purpose choosing inputs targets network that could facilitate reaching reliable Neural Network-based tool challenges associated COVID-19. Furthermore, there some specific each platform, various forms data, such as clinical imaging can improve performance introduced approaches toward best responses practical applications.

Language: Английский

---

Multi-task deep learning based CT imaging analysis for COVID-19 pneumonia: Classification and segmentation

Computers in Biology and Medicine, Journal Year: 2020, Volume and Issue: 126, P. 104037 - 104037

Published: Oct. 8, 2020

Language: Английский

---

A fully automatic deep learning system for COVID-19 diagnostic and prognostic analysis

European Respiratory Journal, Journal Year: 2020, Volume and Issue: 56(2), P. 2000775 - 2000775

Published: May 22, 2020

Coronavirus disease 2019 (COVID-19) has spread globally, and medical resources become insufficient in many regions. Fast diagnosis of COVID-19 finding high-risk patients with worse prognosis for early prevention resource optimisation is important. Here, we proposed a fully automatic deep learning system diagnostic prognostic analysis by routinely used computed tomography. We retrospectively collected 5372 tomography images from seven cities or provinces. Firstly, 4106 were to pre-train the system, making it learn lung features. Following this, 1266 (924 (471 had follow-up >5 days) 342 other pneumonia) six provinces enrolled train externally validate performance system. In four external validation sets, achieved good identifying pneumonia (AUC 0.87 0.88, respectively) viral 0.86). Moreover, succeeded stratify into high- low-risk groups whose hospital-stay time significant difference (p=0.013 p=0.014, respectively). Without human assistance, automatically focused on abnormal areas that showed consistent characteristics reported radiological findings. Deep provides convenient tool fast screening potential patients, which may be helpful before show severe symptoms.

Language: Английский

---

Mapping the landscape of Artificial Intelligence applications against COVID-19

Journal of Artificial Intelligence Research, Journal Year: 2020, Volume and Issue: 69, P. 807 - 845

Published: Nov. 19, 2020

COVID-19, the disease caused by SARS-CoV-2 virus, has been declared a pandemic World Health Organization, which reported over 18 million confirmed cases as of August 5, 2020. In this review, we present an overview recent studies using Machine Learning and, more broadly, Artificial Intelligence, to tackle many aspects COVID-19 crisis. We have identified applications that address challenges posed at different scales, including: molecular, identifying new or existing drugs for treatment; clinical, supporting diagnosis and evaluating prognosis based on medical imaging non-invasive measures; societal, tracking both epidemic accompanying infodemic multiple data sources. also review datasets, tools, resources needed facilitate Intelligence research, discuss strategic considerations related operational implementation multidisciplinary partnerships open science. highlight need international cooperation maximize potential AI in future pandemics.

Language: Английский

---