Estimating Uncertainty and Interpretability in Deep Learning for Coronavirus (COVID-19) Detection

Ghoshal, Biraja; Tucker, Allan

doi:10.48550/arxiv.2003.10769

Cited by 83 publications

(102 citation statements)

References 11 publications

(15 reference statements)

Supporting

Mentioning

102

Contrasting

Order By: Relevance

“…This black-box nature of ViTs has hindered their deployment in clinical practice since, in areas such as medical applications, it is imperative to identify the limitations and potential failure cases of designed systems, where interpretability plays a fundamental role [393]. Although several explainable AIbased medical imaging systems have been developed to gain deeper insights into the working of CNNs models for clinical applications [394]- [396], however, the work is still in its infancy for ViT-based medical imaging applications.…”

Section: Interpretabilitymentioning

confidence: 99%

Transformers in Medical Imaging: A Survey

Shamshad¹,

Khan²,

Zamir³

et al. 2022

Preprint

View full text Add to dashboard Cite

Following unprecedented success on the natural language tasks, Transformers have been successfully applied to several computer vision problems, achieving state-of-the-art results and prompting researchers to reconsider the supremacy of convolutional neural networks (CNNs) as de facto operators. Capitalizing on these advances in computer vision, the medical imaging field has also witnessed growing interest for Transformers that can capture global context compared to CNNs with local receptive fields. Inspired from this transition, in this survey, we attempt to provide a comprehensive review of the applications of Transformers in medical imaging covering various aspects, ranging from recently proposed architectural designs to unsolved issues. Specifically, we survey the use of Transformers in medical image segmentation, detection, classification, reconstruction, synthesis, registration, clinical report generation, and other tasks. In particular, for each of these applications, we develop taxonomy, identify application-specific challenges as well as provide insights to solve them, and highlight recent trends. Further, we provide a critical discussion of the field's current state as a whole, including the identification of key challenges, open problems, and outlining promising future directions. We hope this survey will ignite further interest in the community and provide researchers with an up-to-date reference regarding applications of Transformer models in medical imaging. Finally, to cope with the rapid development in this field, we intend to regularly update the relevant latest papers and their open-source implementations at https://github.com/fahadshamshad/awesome-transformers-in-medical-imaging.

show abstract

Section: Interpretabilitymentioning

confidence: 99%

Transformers in Medical Imaging: A Survey

Shamshad¹,

Khan²,

Zamir³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…A suggested "Corona score" is used to track how patients progress over time. Ghoshal and Tucker (2020) attempted to determine how Bayesian Convolutional Neural Networks (Dropweights based) Networks (BCNN) can calculate uncertainty in DL resolutions to improve the diagnostic accuracy of the human-machine combination using COVID-19 chest X-ray in another paper. Their primary goal is to avoid false-negative detection.…”

Section: Artificial Neural Network In Covid-19mentioning

confidence: 99%

The Prominence of Artificial Intelligence in COVID-19

Nasim¹,

Dhali²,

Afrin³

et al. 2021

Preprint

View full text Add to dashboard Cite

In December 2019, a novel virus called COVID-19 had caused an enormous number of causalities to date. The battle with the novel Coronavirus is baffling and horrifying after the Spanish Flu 2019. While the front-line doctors and medical researchers have made significant progress in controlling the spread of the highly contiguous virus, technology has also proved its significance in the battle. Moreover, Artificial Intelligence has been adopted in many medical applications to diagnose many diseases, even baffling experienced doctors. Therefore, this survey paper explores the methodologies proposed that can aid doctors and researchers in early and inexpensive methods of diagnosis of the disease. Most developing countries have difficulties carrying out tests using the conventional manner, but a significant way can be adopted with Machine and Deep Learning. On the other hand, the access to different types of medical images has motivated the researchers. As a result, a mammoth number of techniques are proposed. This paper first details the background knowledge of the conventional methods in the Artificial Intelligence domain. Following that, we gather the commonly used datasets and their use cases to date. In addition, we also show the percentage of researchers adopting Machine Learning over Deep Learning. Thus we provide a thorough analysis of this scenario. Lastly, in the research challenges, we elaborate on the problems faced in COVID-19 research, and we address the issues with our understanding to build a bright and healthy environment.

show abstract

“…Only a handful of works address the uncertainty of COVID detection methods (Mallick et al 2020;Ghoshal and Tucker 2020). Mallick et.…”

Section: Related Workmentioning

confidence: 99%

Reducing Risk and Uncertainty of Deep Neural Networks on Diagnosing COVID-19 Infection

Sarker¹,

Pandit²,

Sarker³

et al. 2021

Preprint

View full text Add to dashboard Cite

Effective and reliable screening of patients via Computer-Aided Diagnosis can play a crucial part in the battle against COVID-19. Most of the existing works focus on developing sophisticated methods yielding high detection performance, yet not addressing the issue of predictive uncertainty. In this work, we introduce uncertainty estimation to detect confusing cases for expert referral to address the unreliability of stateof-the-art (SOTA) DNNs on COVID-19 detection. To the best of our knowledge, we are the first to address this issue on the COVID-19 detection problem. In this work, we investigate a number of SOTA uncertainty estimation methods on publicly available COVID dataset and present our experimental findings. In collaboration with medical professionals, we further validate the results to ensure the viability of the best performing method in clinical practice.

show abstract

Estimating Uncertainty and Interpretability in Deep Learning for Coronavirus (COVID-19) Detection

Cited by 83 publications

References 11 publications

Transformers in Medical Imaging: A Survey

Transformers in Medical Imaging: A Survey

The Prominence of Artificial Intelligence in COVID-19

Reducing Risk and Uncertainty of Deep Neural Networks on Diagnosing COVID-19 Infection

Contact Info

Product

Resources

About