A Survey on Attention Mechanisms for Medical Applications: are we Moving Toward Better Algorithms? DOI Creative Commons
Tiago Gonçalves, Isabel Rio-Torto, Luís F. Teixeira

et al.

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 98909 - 98935

Published: Jan. 1, 2022

The increasing popularity of attention mechanisms in deep learning algorithms for computer vision and natural language processing made these models attractive to other research domains. In healthcare, there is a strong need tools that may improve the routines clinicians patients. Naturally, use attention-based medical applications occurred smoothly. However, being healthcare domain depends on high-stake decisions, scientific community must ponder if high-performing fit needs applications. With this motto, paper extensively reviews machine methods (including Transformers) several based types tasks integrate works pipelines domain. This work distinguishes itself from its predecessors by proposing critical analysis claims potentialities presented literature through an experimental case study image classification with three different cases. These experiments focus integrating process into established architectures, their predictive power, visual assessment saliency maps generated post-hoc explanation methods. concludes about proposes future lines benefit frameworks.

Language: Английский

Transformers in medical imaging: A survey DOI
Fahad Shamshad, Salman Khan, Syed Waqas Zamir

et al.

Medical Image Analysis, Journal Year: 2023, Volume and Issue: 88, P. 102802 - 102802

Published: April 5, 2023

Language: Английский

Citations

585
A Review of the Role of Artificial Intelligence in Healthcare DOI Open Access
Ahmed Al Kuwaiti,

Khalid Nazer,

Abdullah H. Alreedy

et al.

Journal of Personalized Medicine, Journal Year: 2023, Volume and Issue: 13(6), P. 951 - 951

Published: June 5, 2023

Artificial intelligence (AI) applications have transformed healthcare. This study is based on a general literature review uncovering the role of AI in healthcare and focuses following key aspects: (i) medical imaging diagnostics, (ii) virtual patient care, (iii) research drug discovery, (iv) engagement compliance, (v) rehabilitation, (vi) other administrative applications. The impact observed detecting clinical conditions diagnostic services, controlling outbreak coronavirus disease 2019 (COVID-19) with early diagnosis, providing care using AI-powered tools, managing electronic health records, augmenting compliance treatment plan, reducing workload professionals (HCPs), discovering new drugs vaccines, spotting prescription errors, extensive data storage analysis, technology-assisted rehabilitation. Nevertheless, this science pitch meets several technical, ethical, social challenges, including privacy, safety, right to decide try, costs, information consent, access, efficacy, while integrating into governance crucial for safety accountability raising HCPs' belief enhancing acceptance boosting significant consequences. Effective prerequisite precisely address regulatory, trust issues advancing implementation AI. Since COVID-19 hit global system, concept has created revolution healthcare, such an uprising could be another step forward meet future needs.

Language: Английский

Citations

327
Transformers in medical image analysis DOI Creative Commons
Kelei He, Gan Chen, Zhuoyuan Li

et al.

Intelligent Medicine, Journal Year: 2022, Volume and Issue: 3(1), P. 59 - 78

Published: Aug. 24, 2022

Transformers have dominated the field of natural language processing and recently made an impact in area computer vision. In medical image analysis, transformers also been successfully used to full-stack clinical applications, including synthesis/reconstruction, registration, segmentation, detection, diagnosis. This paper aimed promote awareness applications analysis. Specifically, we first provided overview core concepts attention mechanism built into other basic components. Second, reviewed various transformer architectures tailored for discuss their limitations. Within this review, investigated key challenges use different learning paradigms, improving model efficiency, coupling with techniques. We hope review would provide a comprehensive picture readers interest

Language: Английский

Citations

251
Vision Transformers in medical computer vision—A contemplative retrospection DOI

Arshi Parvaiz,

Muhammad Anwaar Khalid,

Rukhsana Zafar

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 122, P. 106126 - 106126

Published: March 20, 2023

Language: Английский

Citations

146
Advances in medical image analysis with vision Transformers: A comprehensive review DOI
Reza Azad, Amirhossein Kazerouni, Moein Heidari

et al.

Medical Image Analysis, Journal Year: 2023, Volume and Issue: 91, P. 103000 - 103000

Published: Oct. 19, 2023

Language: Английский

Citations

125
Application of explainable artificial intelligence in medical health: A systematic review of interpretability methods DOI Creative Commons
Shahab S. Band,

Atefeh Yarahmadi,

Chung-Chian Hsu

et al.

Informatics in Medicine Unlocked, Journal Year: 2023, Volume and Issue: 40, P. 101286 - 101286

Published: Jan. 1, 2023

This paper investigates the applications of explainable AI (XAI) in healthcare, which aims to provide transparency, fairness, accuracy, generality, and comprehensibility results obtained from ML algorithms decision-making systems. The black box nature systems has remained a challenge interpretable techniques can potentially address this issue. Here we critically review previous studies related interpretability methods medical Descriptions various types XAI such as layer-wise relevance propagation (LRP), Uniform Manifold Approximation Projection (UMAP), Local Interpretable Model-agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP), ANCHOR, contextual importance utility (CIU), Training calibration-based explainers (TraCE), Gradient-weighted Class Activation Mapping (Grad-CAM), t-distributed Stochastic Neighbor Embedding (t-SNE), NeuroXAI, Explainable Cumulative Fuzzy Membership Criterion (X-CFCMC) along with diseases be explained through these are provided throughout paper. also discusses how technologies transform healthcare services. usability reliability presented summarized, including on XGBoost for mediastinal cysts tumors, 3D brain tumor segmentation network, TraCE method image analysis. Overall, contribute growing field insights researchers, practitioners, decision-makers industry. Finally, discuss performance applied health care It is needed mention that brief implemented methodology section.

Language: Английский

Citations

100
Artificial Intelligence of Things for Smarter Healthcare: A Survey of Advancements, Challenges, and Opportunities DOI Creative Commons
Stephanie Baker, Wei Xiang

IEEE Communications Surveys & Tutorials, Journal Year: 2023, Volume and Issue: 25(2), P. 1261 - 1293

Published: Jan. 1, 2023

Healthcare systems are under increasing strain due to a myriad of factors, from steadily ageing global population the current COVID-19 pandemic. In world where we have needed be connected but apart, need for enhanced remote and at-home healthcare has become clear. The Internet Things (IoT) offers promising solution. IoT created highly world, with billions devices collecting communicating data range applications, including healthcare. Due these high volumes data, natural synergy Artificial Intelligence (AI) apparent – big both enables requires AI interpret, understand, make decisions that provide optimal outcomes. this extensive survey, thoroughly explore through an examination field (AIoT) This work begins by briefly establishing unified architecture AIoT in context, sensors devices, novel communication technologies, cross-layer AI. We then examine recent research pertaining each component several key perspectives, identifying challenges, opportunities unique Several examples real-world use cases presented illustrate potential technologies. Lastly, outlines directions future

Language: Английский

Citations

71
Automated detection and forecasting of COVID-19 using deep learning techniques: A review DOI
Afshin Shoeibi, Marjane Khodatars, Mahboobeh Jafari

et al.

Neurocomputing, Journal Year: 2024, Volume and Issue: 577, P. 127317 - 127317

Published: Jan. 26, 2024

Language: Английский

Citations

55
Explainable Vision Transformers and Radiomics for COVID-19 Detection in Chest X-rays DOI Open Access
Mohamed Chetoui, Moulay A. Akhloufi

Journal of Clinical Medicine, Journal Year: 2022, Volume and Issue: 11(11), P. 3013 - 3013

Published: May 26, 2022

The rapid spread of COVID-19 across the globe since its emergence has pushed many countries’ healthcare systems to verge collapse. To restrict disease and lessen ongoing cost on system, it is critical appropriately identify COVID-19-positive individuals isolate them as soon possible. primary screening test, RT-PCR, although accurate reliable, a long turn-around time. More recently, various researchers have demonstrated use deep learning approaches chest X-ray (CXR) for detection. However, existing Deep Convolutional Neural Network (CNN) methods fail capture global context due their inherent image-specific inductive bias. In this article, we investigated vision transformers (ViT) detecting in Chest images. Several ViT models were fine-tuned multiclass classification problem (COVID-19, Pneumonia Normal cases). A dataset consisting 7598 CXR images, 8552 healthy patients 5674 used. obtained results achieved high performance with an Area Under Curve (AUC) 0.99 multi-class (COVID-19 vs. Other normal). sensitivity class 0.99. We that outperformed comparable state-of-the-art images using CNN architectures. attention map proposed model showed our able efficiently signs COVID-19.

Language: Английский

Citations

41
Deep learning framework for rapid and accurate respiratory COVID-19 prediction using chest X-ray images DOI Creative Commons
Chiagoziem C. Ukwuoma, Dongsheng Cai, Md Belal Bin Heyat

et al.

Journal of King Saud University - Computer and Information Sciences, Journal Year: 2023, Volume and Issue: 35(7), P. 101596 - 101596

Published: May 25, 2023

COVID-19 is a contagious disease that affects the human respiratory system. Infected individuals may develop serious illnesses, and complications result in death. Using medical images to detect from essentially identical thoracic anomalies challenging because it time-consuming, laborious, prone error. This study proposes an end-to-end deep-learning framework based on deep feature concatenation Multi-head Self-attention network. Feature involves fine-tuning pre-trained backbone models of DenseNet, VGG-16, InceptionV3, which are trained large-scale ImageNet, whereas network adopted for performance gain. End-to-end training evaluation procedures conducted using COVID-19_Radiography_Dataset binary multi-classification scenarios. The proposed model achieved overall accuracies (96.33% 98.67%) F1_scores (92.68% multi classification scenarios, respectively. In addition, this highlights difference accuracy (98.0% vs. 96.33%) F_1 score (97.34% 95.10%) when compared with against highest individual performance. Furthermore, virtual representation saliency maps employed attention mechanism focusing abnormal regions presented explainable artificial intelligence (XAI) technology. provided better prediction results outperforming other recent learning same dataset.

Language: Английский

Citations

39