Performance of an AI algorithm during the different phases of the COVID pandemics: what can we learn from the AI and vice versa.

Catalano, Michele; Bortolotto, Chandra; Nicora, Giovanna; Achilli, Marina Francesca; Consonni, Alessio; Ruongo, Lidia; Callea, Giovanni; Lo Tito, Antonio; Biasibetti, Carla; Donatelli, Antonella; Cutti, Sara; Comotto, Federico; Stella, Giulia Maria; Corsico, Angelo; Perlini, Stefano; Bellazzi, Riccardo; Bruno, Raffaele; Filippi, Andrea; Preda, Lorenzo

doi:10.1016/j.ejro.2023.100497

Cited by 6 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As future lines of studies, larger projects integrating diverse imaging modalities (such as ultrasound and magnetic resonance) also across other parts of the human body, hold promise in enhancing the applicability of minimally autopsy methods, not only in fatal COVID-19. Additionally, the incorporation of artificial intelligence and machine learning in the analysis of postmortem CT images, as it is already being used in living patients [24] , [27] , has the potential to significantly enhance the precision and interpretation of imaging, especially if a large histopathological dataset is input.…”

Section: Discussionmentioning

confidence: 99%

Postmortem chest computed tomography in COVID-19: A minimally invasive autopsy method

Savoia,

Valente Yamada Sawamura,

de Almeida Monteiro

et al. 2024

European Journal of Radiology Open

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Postmortem chest computed tomography in COVID-19: A minimally invasive autopsy method

Savoia,

Valente Yamada Sawamura,

de Almeida Monteiro

et al. 2024

European Journal of Radiology Open

View full text Add to dashboard Cite

“…During the "third wave" (from March to May 2021), 462 additional patients were triaged in our Emergency Department, and 68% of them were hospitalized. This dataset is based on the ALFABETO (ALl FAster BEtter TOgether) project, whose aim is to develop an AI-based pipeline integrating data from diagnostic tools and clinical features to support clinicians during the triage of COVID-19 patients [31]. The third wave set was exploited as an additional, independent test validation set.…”

Section: Methodsmentioning

confidence: 99%

Bayesian Networks in the Management of Hospital Admissions: A Comparison between Explainable AI and Black Box AI during the Pandemic

Nicora,

Catalano,

Bortolotto

et al. 2024

J. Imaging

Self Cite

View full text Add to dashboard Cite

Artificial Intelligence (AI) and Machine Learning (ML) approaches that could learn from large data sources have been identified as useful tools to support clinicians in their decisional process; AI and ML implementations have had a rapid acceleration during the recent COVID-19 pandemic. However, many ML classifiers are “black box” to the final user, since their underlying reasoning process is often obscure. Additionally, the performance of such models suffers from poor generalization ability in the presence of dataset shifts. Here, we present a comparison between an explainable-by-design (“white box”) model (Bayesian Network (BN)) versus a black box model (Random Forest), both studied with the aim of supporting clinicians of Policlinico San Matteo University Hospital in Pavia (Italy) during the triage of COVID-19 patients. Our aim is to evaluate whether the BN predictive performances are comparable with those of a widely used but less explainable ML model such as Random Forest and to test the generalization ability of the ML models across different waves of the pandemic.

show abstract

“…No statistical test showed whether the betterment had been significant. (iii) Another study [77] examined the "discrepancies" between AI suggestions and clinicians' actual decisions on whether the patients should be treated in a spoke-or a hub center. Here, five parameters were considered: accuracy (76%), AUC ROC (83%), specificity (78%), recall (74%), and precision, i.e., positive predicted value (88%).…”

Section: Other Approaches (Outlook Perspective)mentioning

confidence: 99%

“…Here, five parameters were considered: accuracy (76%), AUC ROC (83%), specificity (78%), recall (74%), and precision, i.e., positive predicted value (88%). The enigmatic formulation of conclusions-"[the code] is in line or slightly worse than (iii) Another study [77] examined the "discrepancies" between AI suggestions and clinicians' actual decisions on whether the patients should be treated in a spoke-or a hub center. Here, five parameters were considered: accuracy (76%), AUC ROC (83%), specificity (78%), recall (74%), and precision, i.e., positive predicted value (88%).…”

Section: Other Approaches (Outlook Perspective)mentioning

confidence: 99%

Frequent Errors in Modeling by Machine Learning: A Prototype Case of Predicting the Timely Evolution of COVID-19 Pandemic

Héberger

2024

Algorithms

View full text Add to dashboard Cite

Background: The development and application of machine learning (ML) methods have become so fast that almost nobody can follow their developments in every detail. It is no wonder that numerous errors and inconsistencies in their usage have also spread with a similar speed independently from the tasks: regression and classification. This work summarizes frequent errors committed by certain authors with the aim of helping scientists to avoid them. Methods: The principle of parsimony governs the train of thought. Fair method comparison can be completed with multicriteria decision-making techniques, preferably by the sum of ranking differences (SRD). Its coupling with analysis of variance (ANOVA) decomposes the effects of several factors. Earlier findings are summarized in a review-like manner: the abuse of the correlation coefficient and proper practices for model discrimination are also outlined. Results: Using an illustrative example, the correct practice and the methodology are summarized as guidelines for model discrimination, and for minimizing the prediction errors. The following factors are all prerequisites for successful modeling: proper data preprocessing, statistical tests, suitable performance parameters, appropriate degrees of freedom, fair comparison of models, and outlier detection, just to name a few. A checklist is provided in a tutorial manner on how to present ML modeling properly. The advocated practices are reviewed shortly in the discussion. Conclusions: Many of the errors can easily be filtered out with careful reviewing. Every authors’ responsibility is to adhere to the rules of modeling and validation. A representative sampling of recent literature outlines correct practices and emphasizes that no error-free publication exists.

show abstract

Performance of an AI algorithm during the different phases of the COVID pandemics: what can we learn from the AI and vice versa.

Cited by 6 publications

References 28 publications

Postmortem chest computed tomography in COVID-19: A minimally invasive autopsy method

Postmortem chest computed tomography in COVID-19: A minimally invasive autopsy method

Bayesian Networks in the Management of Hospital Admissions: A Comparison between Explainable AI and Black Box AI during the Pandemic

Frequent Errors in Modeling by Machine Learning: A Prototype Case of Predicting the Timely Evolution of COVID-19 Pandemic

Contact Info

Product

Resources

About