2020
DOI: 10.1097/ccm.0000000000004468
|View full text |Cite
|
Sign up to set email alerts
|

External Validation of an Acute Respiratory Distress Syndrome Prediction Model Using Radiology Reports

Abstract: Objectives: Acute respiratory distress syndrome is frequently under recognized and associated with increased mortality. Previously, we developed a model that used machine learning and natural language processing of text from radiology reports to identify acute respiratory distress syndrome. The model showed improved performance in diagnosing acute respiratory distress syndrome when compared to a rule-based method. In this study, our objective was to externally validate the natural language processi… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
8
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
6

Relationship

0
6

Authors

Journals

citations
Cited by 9 publications
(8 citation statements)
references
References 23 publications
0
8
0
Order By: Relevance
“…This retrospective cohort study included adult ARDS patients admitted between 2017 and 2021 to a 23-hospital system in the Intermountain West. We abstracted data from the electronic health record and used natural language processing to identify radiographic pneumothorax and/or pneumomediastinum [ 3 , 4 ]. We performed bivariate and adjusted analyses to compare patients with pre-pandemic ARDS (2017–2020) to patients with a positive SARS-CoV-2 polymerase chain reaction (PCR) result proximate to ARDS (2020–2021) (see also Supplemental Methods).…”
mentioning
confidence: 99%
See 1 more Smart Citation
“…This retrospective cohort study included adult ARDS patients admitted between 2017 and 2021 to a 23-hospital system in the Intermountain West. We abstracted data from the electronic health record and used natural language processing to identify radiographic pneumothorax and/or pneumomediastinum [ 3 , 4 ]. We performed bivariate and adjusted analyses to compare patients with pre-pandemic ARDS (2017–2020) to patients with a positive SARS-CoV-2 polymerase chain reaction (PCR) result proximate to ARDS (2020–2021) (see also Supplemental Methods).…”
mentioning
confidence: 99%
“…We investigated whether COVID-19 ARDS is associated with more radiographic pneumothorax and/or pneumomediastinum than pre-pandemic ARDS and whether pneumothorax/ pneumomediastinum in COVID-19 ARDS is associated with worse outcomes or differing treatments.This retrospective cohort study included adult ARDS patients admitted between 2017 and 2021 to a 23-hospital system in the Intermountain West. We abstracted data from the electronic health record and used natural language processing to identify radiographic pneumothorax and/or pneumomediastinum [3,4]. We performed bivariate and adjusted analyses to compare patients with pre-pandemic ARDS (2017-2020) to patients with a positive SARS-CoV-2 polymerase chain reaction (PCR) result proximate to ARDS (2020-2021) (see also Supplemental Methods).Comparing 2,211 patients with COVID-19 ARDS and 5522 with pre-pandemic ARDS (Table 1 and Supplemental Fig.…”
mentioning
confidence: 99%
“…Machine learning models to predict (as opposed to diagnose) PARDS could facilitate even earlier identification of patients at risk than current systems. Although there are examples of these types of models in the literature (17–20) , further research is needed, including whether patients who are identified early by these models will ultimately have a net clinical benefit through earlier interventions.…”
Section: Resultsmentioning
confidence: 99%
“…Although there are examples of these types of models in the literature (17)(18)(19)(20), further research is needed, including whether patients who are identified early by these models will ultimately have a net clinical benefit through earlier interventions.…”
Section: S5mentioning
confidence: 99%
“…13,76 In supervised machine learning, the outcome from machine judgment is constantly prognostication of ARDS/ALI. [77][78][79] Machine learning algorithms have also been successfully used to estimate lung mechanics during mechanical ventilation. 80 In using machine learning techniques, investigators need to have clinical expertise and the ability to understand machine learning algorithms.…”
Section: Ai For Clinical Studies Involving Ards/alimentioning
confidence: 99%