ABSTRACT
The most promising standard machine learning methods can deliver highly accurate classification results, often outperforming standard white-box methods. However, it is hardly possible for humans to fully understand the rationale behind the black-box results, and thus, these powerful methods hamper the creation of new knowledge on the part of humans and the broader acceptance of this technology. Explainable Artificial Intelligence attempts to overcome this problem by making the results more interpretable, while Interactive Machine Learning integrates humans into the process of insight discovery. The paper builds on recent successes in combining these two cutting-edge technologies and proposes how Explanatory Interactive Machine Learning (XIL) is embedded in a generalizable Action Design Research (ADR) process – called XIL-ADR. This approach can be used to analyze data, inspect models, and iteratively improve them. The paper shows the application of this process using the diagnosis of viral pneumonia, e.g., Covid-19, as an illustrative example. By these means, the paper also illustrates how XIL-ADR can help identify shortcomings of standard machine learning projects, gain new insights on the part of the human user, and thereby can help to unlock the full potential of AI-based systems for organizations and research. © 2023, The Author(s).
ABSTRACT
Background: The coronavirus disease 2019 (COVID-19) has resulted in an unprecedented research response, demonstrating exceptional examples of rapid research and collaboration. There has however been an ongoing need for greater coordination, with limited resources for research and the shifting global pandemic.
ABSTRACT
Background: The coronavirus disease 2019 (COVID-19) has resulted in an unprecedented research response, demonstrating exceptional examples of rapid research and collaboration. There has however been an ongoing need for greater coordination, with limited resources for research and the shifting global nature of the pandemics.
ABSTRACT
Background: The coronavirus disease 2019 (COVID-19) has resulted in an unprecedented research response, demonstrating exceptional examples of rapid research and collaboration. There is however a need for greater coordination, with limited resources and the shifting global nature of the pandemic resulting in a proliferation of research projects underpowered and unable to achieve their aims. Methods: The UK Collaborative on Development Research (UKCDR) and Global Research Collaboration for Infectious Disease Preparedness (GloPID-R), two funder coordination groups have collaborated to develop a live database of funded research projects across the world relating to COVID-19. Drawing data continually from their members and further global funding bodies, as of 15 (th) July 2021 the database contains 12,419 projects, funded by 255 funders, taking place across 149 countries representing an investment of at least $4.9 billion. To our knowledge it is one of the most comprehensive databases. The database is aligned to the World Health Organisation and GloPID-R Global Research Roadmap: 2019 Novel Coronavirus. It is being used by the WHO, governments and multi-lateral policy makers, research funders and researchers. This living mapping review aims to supplement the database by providing an open accessible and frequently updated resource summarising the characteristics of the COVID-19 funded research portfolio. Both descriptive and thematic analysis will be presented and updated frequently to aid interpretation of the global COVID-19 funded research portfolio. Results: In this version five analysis we provide an updated detailed descriptive analysis of the database (three months after version four) and focus our thematic analysis on research gaps, research areas in need of coordination, study populations and research locations (with a focus on resource-limited countries). Conclusions: As the global funding response to COVID-19 plateaus, this living mapping review helps both funders and researchers to prioritise resources to areas where there is continued unmet research need.
ABSTRACT
Automatic segmentation of lung lesions in computer tomography has the potential to ease the burden of clinicians during the Covid-19 pandemic. Yet predictive deep learning models are not trusted in the clinical routine due to failing silently in out-of-distribution (OOD) data. We propose a lightweight OOD detection method that exploits the Mahalanobis distance in the feature space. The proposed approach can be seamlessly integrated into state-of-the-art segmentation pipelines without requiring changes in model architecture or training procedure, and can therefore be used to assess the suitability of pre-trained models to new data. We validate our method with a patch-based nnU-Net architecture trained with a multi-institutional dataset and find that it effectively detects samples that the model segments incorrectly. © 2021, Springer Nature Switzerland AG.
ABSTRACT
Background: The coronavirus disease 2019 (COVID-19) has resulted in an unprecedented research response, demonstrating exceptional examples of rapid research and collaboration. There is however a need for greater coordination, with limited resources and the shifting global nature of the pandemic resulting in a proliferation of research projects underpowered and unable to achieve their aims.
ABSTRACT
Background: The coronavirus disease 2019 (COVID-19) has resulted in an unprecedented research response, demonstrating exceptional examples of rapid research and collaboration. There is however a need for greater coordination, with limited resources and the shifting global nature of the pandemic resulting in a proliferation of research projects underpowered and unable to achieve their aims.