Radiology January 2021 PMID 32787701

Model-based Prediction of Critical Illness in Hospitalized Patients with COVID-19.

Steven Schalekamp, Merel Huisman, Rogier A van Dijk, Martijn F Boomsma, Pedro J Freire Jorge, Wytze S de Boer, Gerada Judith M Herder, Marja Bonarius, Oscar A Groot, Eefje Jong, Anton Schreuder, Cornelia M Schaefer-Prokop

Full text PDF Download social card

Mentions

Peak: 28 in Aug 2020

across 3 sources since Aug 2020

Extended reach

283,490

potential audience

Total likes

185

Video views

11,616

Mentions per month

Aug 2020 – Jan 2021

Where it's discussed

X / Twitter30 posts
YouTube2 videos
Facebook2 posts

AI synopsis

What people are saying

The discussion surrounding the article reflects a positive reception, with phrases like 'hot off the press' and mentions of 'new risk calculators,' emphasizing recent developments. Several posts bring up patient cases and practical implications, using words like 'could' and 'possible,' which suggest genuine curiosity and acknowledgment of the model's potential to influence clinical practice. The tone is generally professional and focused on the model's usefulness, but there's limited critical engagement, indicating mild to moderate depth of discussion.

AgreementModerate agreement

Most discussions acknowledge and support the importance of the prediction model, with some posts highlighting the publication as timely and relevant during the pandemic.

InterestHigh interest

The posts demonstrate high interest, especially through references to new models, risk calculators, and specific patient scenarios, indicating curiosity about practical applications.

EngagementModerate engagement

Engagement appears moderate, with some posts sharing summaries, reactions, and citations, but limited deep analysis or critique of the methodology.

ImpactModerate impact

While there is recognition of the potential impact on clinical decision-making, many comments are more informative or promotional rather than explicitly assessing the model's broader influence.

Abstract in brief

The study developed a prediction model using clinical, radiographic, and laboratory data at hospital admission to identify COVID-19 patients at risk of critical illness, such as death or needing intensive care with invasive ventilation.
The model showed good predictive performance with an area under the curve (AUC) of 0.77, indicating it can reasonably distinguish between patients who may or may not develop severe illness.
This tool can help healthcare providers assess patient risk early on, potentially optimizing ICU resource allocation during pandemic situations.

Read the full abstract

All mentions

Advances in Chest Imaging and AI for COVID-19 Severity Assessment

Following the April 2020 publication of the Fleischner Statement, Dr. Geoffrey Rubin provides an update based on five months of new evidence. The findings emphasize the limited role of CT in diagnosis but highlight its…

RSNA · Sep 21, 2020 11,286 views

Predicting Critical Illness in COVID-19 Patients Using Model-Based Approaches

This study developed a prediction model using clinical, radiographic, and laboratory data at hospital admission to identify COVID-19 patients at risk of critical illness, such as death or needing intensive care with…

RadiologyRSNA · Aug 26, 2020 330 views

Reira Espinosa

@reira_espinosa · X

RT @e_Radiologia: [Materiais mais acessados do e-Radiologia na semana #01] 1. [Original Research] Model-based Prediction of Critical Illne…

Jan 11, 2021 · 1 share · 186 followers View post →

e-Radiologia

@e_Radiologia · X

Jan 11, 2021 · 2 likes · 1 share · 4,727 followers View post →

e-Radiologia

@e_Radiologia · X

[Original Research] Model-based Prediction of Critical Illness in Hospitalized Patients with COVID-19 - https://t.co/MO5vb6YN8t https://t.co/Nix1j37q22

Jan 7, 2021 · 4,727 followers View post →