Pre-hospital prediction of adverse outcomes in patients with suspected COVID-19: Development, application and comparison of machine learning and deep learning methods

Hasan, Madina; Bath, Peter A.; Marincowitz, Carl; Sutton, Laura; Pilbery, Richard; Hopfgartner, Frank; Mazumdar, Suvodeep; Campbell, R.; Stone, Tony; Thomas, Ben; Bell, Fiona; Turner, Janette; Biggs, Katie; Petrie, Jennifer; Goodacre, Steve

doi:10.1016/j.compbiomed.2022.106024

Cited by 11 publications

(7 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A complete list of detailed training (as well as interpretation) methods used in the identification phase of the literature in our paper is provided in the Supplementary Material to this paper. Among statistical regression models, the logistic regression (LR) model is frequently used to take on either classification or regression tasks in various studies [93][94][95]. Model training in the context of natural language processing and medical imaging is often elaborated through pre-trained models, such as Bert [44], ResNet [96], etc.…”

Section: Model Trainingmentioning

confidence: 99%

“…Ensemble models incorporate a number of basic ML or DL models to achieve higher degrees of accuracy. Four main alternatives to creating ensembles, comprising bagging, boosting, stacking, and mixture of experts, are addressed in [93]. Hybrid models concatenate different combinations of ML and DL models at different model architecture levels [96,101,102].…”

Section: Model Trainingmentioning

confidence: 99%

See 1 more Smart Citation

Towards Improved XAI-Based Epidemiological Research into the Next Potential Pandemic

Khalili,

Wimmer

2024

Life

View full text Add to dashboard Cite

By applying AI techniques to a variety of pandemic-relevant data, artificial intelligence (AI) has substantially supported the control of the spread of the SARS-CoV-2 virus. Along with this, epidemiological machine learning studies of SARS-CoV-2 have been frequently published. While these models can be perceived as precise and policy-relevant to guide governments towards optimal containment policies, their black box nature can hamper building trust and relying confidently on the prescriptions proposed. This paper focuses on interpretable AI-based epidemiological models in the context of the recent SARS-CoV-2 pandemic. We systematically review existing studies, which jointly incorporate AI, SARS-CoV-2 epidemiology, and explainable AI approaches (XAI). First, we propose a conceptual framework by synthesizing the main methodological features of the existing AI pipelines of SARS-CoV-2. Upon the proposed conceptual framework and by analyzing the selected epidemiological studies, we reflect on current research gaps in epidemiological AI toolboxes and how to fill these gaps to generate enhanced policy support in the next potential pandemic.

show abstract

Section: Model Trainingmentioning

confidence: 99%

Section: Model Trainingmentioning

confidence: 99%

Towards Improved XAI-Based Epidemiological Research into the Next Potential Pandemic

Khalili,

Wimmer

2024

Life

View full text Add to dashboard Cite

show abstract

“…By analyzing patient data such as vital signs, medical history, and demographic information, these models can provide early warnings to EMS personnel, allowing them to intervene quickly and potentially prevent a critical event from occurring. Other applications of prehospital AI include automated triage, diagnosis support, and resource allocation optimization [113][114][115][116][117][118][119][120][121][122]. Table 3 provides a comprehensive overview of clinical research studies conducted in prehospital care using AI, with a focus on their findings and implications for practice [123][124][125][126][127][128][129].…”

Section: What Are the Future Opportunities And Perspectives?mentioning

confidence: 99%

Clinical Research in Prehospital Care: Current and Future Challenges

Cimino,

Braun

2023

Clinics and Practice

View full text Add to dashboard Cite

Prehospital care plays a critical role in improving patient outcomes, particularly in cases of time-sensitive emergencies such as trauma, cardiac failure, stroke, bleeding, breathing difficulties, systemic infections, etc. In recent years, there has been a growing interest in clinical research in prehospital care, and several challenges and opportunities have emerged. There is an urgent need to adapt clinical research methodology to a context of prehospital care. At the same time, there are many barriers in prehospital research due to the complex context, posing unique challenges for research, development, and evaluation. Among these, this review allows the highlighting of limited resources and infrastructure, ethical and regulatory considerations, time constraints, privacy, safety concerns, data collection and analysis, selection of a homogeneous study group, etc. The analysis of the literature also highlights solutions such as strong collaboration between emergency medical services (EMS) and hospital care, use of (mobile) health technologies and artificial intelligence, use of standardized protocols and guidelines, etc. Overall, the purpose of this narrative review is to examine the current state of clinical research in prehospital care and identify gaps in knowledge, including the challenges and opportunities for future research.

show abstract

“…Stacked ensembles have proven to generally be more accurate prediction models than any one base learner alone in clinical contexts [12] , [13] , [14] . In particular, a large number of studies have used stacked ensembles to study COVID-19 data, with many of them focusing on mortality (e.g., [15] , [16] , [17] , [18] , [19] , [20] , [21] ) and a few assessing cardiac events [22] , [23] . In spite of this progress, it remains unclear how to define the best model combinations for strong performance when using stacked generalization.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of stacked ensemble model performance to predict clinical outcomes: A COVID-19 study

Kablan,

Miller,

Suliman

et al. 2023

International Journal of Medical Informatics

View full text Add to dashboard Cite

Pre-hospital prediction of adverse outcomes in patients with suspected COVID-19: Development, application and comparison of machine learning and deep learning methods

Cited by 11 publications

References 41 publications

Towards Improved XAI-Based Epidemiological Research into the Next Potential Pandemic

Towards Improved XAI-Based Epidemiological Research into the Next Potential Pandemic

Clinical Research in Prehospital Care: Current and Future Challenges

Evaluation of stacked ensemble model performance to predict clinical outcomes: A COVID-19 study

Contact Info

Product

Resources

About