StynMedGAN: Medical images augmentation using a new GAN model for improved diagnosis of diseases DOI
Aamir Wali, Muzammil Ahmad, Asma Naseer

et al.

Journal of Intelligent & Fuzzy Systems, Journal Year: 2023, Volume and Issue: 44(6), P. 10027 - 10044

Published: April 4, 2023

Deep networks require a considerable amount of training data otherwise these generalize poorly. Data Augmentation techniques help the network better by providing more variety in data. Standard augmentation such as flipping, and scaling, produce new that is modified version original Generative Adversarial (GANs) have been designed to generate can be exploited. In this paper, we propose GAN model, named StynMedGAN for synthetically generating medical images improve performance classification models. builds upon state-of-the-art styleGANv2 has produced remarkable results all kinds natural images. We introduce regularization term normalized loss factor existing discriminator styleGANv2. It used force generator penalize it if fails. Medical imaging modalities, X-Rays, CT-Scans, MRIs are different nature, show proposed extends capacity handle way. This model (StynMedGAN) applied three types imaging: CT scans, MRI tasks. To validate effectiveness classification, 3 classifiers (CNN, DenseNet121, VGG-16) used. Results trained with StynMedGAN-augmented outperform other methods only The achieved 100%, 99.6%, 100% chest X-Ray, Chest Brain respectively. promising favor potentially important resource practitioners radiologists diagnose diseases.

Language: Английский

Review of COVID-19 testing and diagnostic methods DOI Creative Commons
Olena Filchakova,

Dina Dossym,

Aisha Ilyas

et al.

Talanta, Journal Year: 2022, Volume and Issue: 244, P. 123409 - 123409

Published: April 1, 2022

More than six billion tests for COVID-19 has been already performed in the world. The testing SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus-2) virus and corresponding human antibodies is essential not only diagnostics treatment of infection by medical institutions, but also as a pre-requisite major semi-normal economic social activities such international flights, off line work study offices, access to malls, sport events. Accuracy, sensitivity, specificity, time results cost per test are parameters those even minimal improvement any them may have noticeable impact on life many countries We described, analyzed compared methods detection, while representing their 22 tables. Also, we performance some FDA approved kits with clinical non-FDA just described scientific literature. RT-PCR still remains golden standard detection virus, pressing need alternative less expensive, more rapid, point care evident. Those that eventually get developed satisfy this explained, discussed, quantitatively compared. review bioanalytical chemistry prospective, it be interesting broader circle readers who interested understanding testing, helping leave pandemic past.

Language: Английский

Citations

182
Supply chain risk management with machine learning technology: A literature review and future research directions DOI Creative Commons
Mei Yang, Ming K. Lim, Yingchi Qu

et al.

Computers & Industrial Engineering, Journal Year: 2022, Volume and Issue: 175, P. 108859 - 108859

Published: Dec. 2, 2022

Coronavirus disease 2019 (COVID-19) has placed tremendous pressure on supply chain risk management (SCRM) worldwide. Recent technological advances, especially machine learning (ML) technology, have shown the possibility to prevent (SCR) by decreasing need for human labor, increasing response speed, and predicting risk. However, literature lacks a comprehensive analysis of relationship between ML SCRM. This work conducts review relatively limited in this field. An 67 shortlisted articles from 9 databases shows that area is still rapid development stage researchers extraordinary interest it. The main purpose study current research status so clear understanding gaps area. Moreover, provides an opportunity practitioners pay attention algorithms SCRM during COVID-19 pandemic.

Language: Английский

Citations

80
Detection of Covid-19 and other pneumonia cases from CT and X-ray chest images using deep learning based on feature reuse residual block and depthwise dilated convolutions neural network DOI Open Access
Gaffari Çelik

Applied Soft Computing, Journal Year: 2022, Volume and Issue: 133, P. 109906 - 109906

Published: Dec. 7, 2022

Language: Английский

Citations

59
Class-Aware Adversarial Transformers for Medical Image Segmentation DOI Creative Commons
Chenyu You, Ruihan Zhao, Fenglin Liu

et al.

arXiv (Cornell University), Journal Year: 2022, Volume and Issue: unknown

Published: Jan. 1, 2022

Transformers have made remarkable progress towards modeling long-range dependencies within the medical image analysis domain. However, current transformer-based models suffer from several disadvantages: (1) existing methods fail to capture important features of images due naive tokenization scheme; (2) information loss because they only consider single-scale feature representations; and (3) segmentation label maps generated by are not accurate enough without considering rich semantic contexts anatomical textures. In this work, we present CASTformer, a novel type adversarial transformers, for 2D segmentation. First, take advantage pyramid structure construct multi-scale representations handle variations. We then design class-aware transformer module better learn discriminative regions objects with structures. Lastly, utilize an training strategy that boosts accuracy correspondingly allows discriminator high-level semantically correlated contents low-level features. Our experiments demonstrate CASTformer dramatically outperforms previous state-of-the-art approaches on three benchmarks, obtaining 2.54%-5.88% absolute improvements in Dice over models. Further qualitative provide more detailed picture model's inner workings, shed light challenges improved transparency, transfer learning can greatly improve performance reduce size datasets training, making strong starting point downstream tasks.

Language: Английский

Citations

58
Automated multi-class classification of lung diseases from CXR-images using pre-trained convolutional neural networks DOI
Sahebgoud Hanamantray Karaddi, Lakhan Dev Sharma

Expert Systems with Applications, Journal Year: 2022, Volume and Issue: 211, P. 118650 - 118650

Published: Aug. 27, 2022

Language: Английский

Citations

55
Automatic detection of Covid-19 from chest X-ray and lung computed tomography images using deep neural networks and transfer learning DOI Open Access
Linh T. Duong, Phuong T. Nguyen, Ludovico Iovino

et al.

Applied Soft Computing, Journal Year: 2022, Volume and Issue: 132, P. 109851 - 109851

Published: Nov. 25, 2022

Language: Английский

Citations

46
Machine learning empowered COVID-19 patient monitoring using non-contact sensing: An extensive review DOI Creative Commons
Umer Saeed, Syed Yaseen Shah, Jawad Ahmad

et al.

Journal of Pharmaceutical Analysis, Journal Year: 2022, Volume and Issue: 12(2), P. 193 - 204

Published: Jan. 4, 2022

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which caused the disease 2019 (COVID-19) pandemic, has affected more than 400 million people worldwide. With recent rise of new Delta and Omicron variants, efficacy vaccines become an important question. goal various studies been to limit spread virus by utilizing wireless sensing technologies prevent human-to-human interactions, particularly for healthcare workers. In this paper, we discuss current literature on invasive/contact non-invasive/non-contact (including Wi-Fi, radar, software-defined radio) that have effectively used detect, diagnose, monitor human activities COVID-19 related symptoms, such as irregular respiration. addition, focused cutting-edge machine learning algorithms (such generative adversarial networks, random forest, multilayer perceptron, support vector machine, extremely randomized trees, k-nearest neighbors) their essential role in intelligent systems. Furthermore, study highlights limitations non-invasive techniques prospective research directions.

Language: Английский

Citations

45
Multi-classification deep CNN model for diagnosing COVID-19 using iterative neighborhood component analysis and iterative ReliefF feature selection techniques with X-ray images DOI

Narin Aslan,

Gonca Özmen Koca, Mehmet Ali Kobat

et al.

Chemometrics and Intelligent Laboratory Systems, Journal Year: 2022, Volume and Issue: 224, P. 104539 - 104539

Published: March 30, 2022

Language: Английский

Citations

39
COVID-19 detection from chest CT images using optimized deep features and ensemble classification DOI Creative Commons
Muhammad Minoar Hossain, Md. Abul Ala Walid, S. M. Saklain Galib

et al.

Systems and Soft Computing, Journal Year: 2024, Volume and Issue: 6, P. 200077 - 200077

Published: Feb. 4, 2024

Diagnosis of COVID-19 positive patients is the eventual move to impede expansion coronavirus. Variations coronavirus make it tough recognize through symptoms. Hence, this research aims at a faster and automatic detection approach disease from chest Computed tomography (CT) scan images. For composition system, constructs feature vector CT images features fusion two Convolutional neural network (CNN) models namely VGG-19 ResNet-50. Before fusion, preprocessing techniques are applied gain more accurate outcomes. Moreover, pertinent identified by using several optimization methods Recursive elimination (RFE), Principal component analysis (PCA), Linear discriminant (LDA), among them, we have observed PCA as best preference. Classification performed on optimized utilizing Max voting ensemble classification (MVEC). The fused ResNet-50, processed with MVEC, provide outcomes accuracy, specificity, sensitivity, precision 98.51%, 97.58%, 99.49%, 97.47%, respectively, after 5-fold cross-validation for proposed method.

Language: Английский

Citations

11
Advances of AI in image-based computer-aided diagnosis: A review DOI Creative Commons
Mst. Nilufa Yeasmin, Md Al Amin,

Tasmim Jamal Joti

et al.

Array, Journal Year: 2024, Volume and Issue: 23, P. 100357 - 100357

Published: July 6, 2024

Over the past two decades, computer-aided detection and diagnosis have emerged as a field of research. The primary goal is to enhance diagnostic treatment procedures for radiologists clinicians in medical image analysis. With help big data advanced artificial intelligence (AI) technologies, such machine learning deep algorithms, healthcare system can be made more convenient, active, efficient, personalized. this literature survey was present thorough overview most important developments related (CAD) systems imaging. This considerable importance researchers professionals both computer sciences. Several reviews on specific facets CAD imaging been published. Nevertheless, main emphasis study cover complete range capabilities review article introduces background concepts used typical by outlining comparing several methods frequently employed recent studies. also presents comprehensive well-structured medicine, drawing meticulous selection relevant publications. Moreover, it describes process handling images state-of-the-art AI-based technologies imaging, along with future directions CAD. indicates that algorithms are effective method diagnose detect diseases.

Language: Английский

Citations

11