Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies · 2025
Child and Family Outcomes After PICU Admission: Creation of an Open Access Literature Database Using a Global Team of PICU Specialists Integrated With Machine Learning.
СтатьяScoping ReviewНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Hay Rebecca E, Zorko David J, O'Hearn Katie, McQuaid Cara, Thibodeau Geneviève Du Pont, Garcia Guerra Gonzalo, Olivier Jeremy, Ducharme-Crevier Laurence, Lee Laurie, Del Bel Michael J, Choong Karen, McNally James Dayre
Аннотация
OBJECTIVES: As research examining child health outcomes after PICU admission grows, so does the need for the identification and synthesis of a large body of literature. We aimed to create an open-access scoping repository of literature describing longer-term health outcomes after PICU admission, using a large multinational team (crowdsourcing) and a machine learning (ML) algorithm.
DATA SOURCES: We performed a registered scoping review (OSF DOI10.17605/OSF.IO/HE5VB; Registered November 21, 2022) using MEDLINE, Embase, CINAHL, and CENTRAL databases, 2000-2022, with no language restrictions.
STUDY SELECTION: Observational or interventional studies describing outcomes of children (0-17 yr old) and their families or caregivers measured greater than 2 weeks post-PICU discharge. Titles and abstracts and full texts were initially screened by a large team of PICU healthcare workers and researchers who were recruited as part of an Evidence Hackathon event at the 2022 World Federation of Pediatric Intensive and Critical Care Societies conference. Initial screening results from 5000 citations were used to develop and validate an ML algorithm, after which a hybrid human crowdsourcing and ML approach was used to screen the remaining 11,055 studies.
DATA EXTRACTION: Not applicable.
DATA SYNTHESIS: Of 16,055 eligible citations, 1,301 met the criteria at full text for inclusion in the database. The screening was completed in just under 2 months while adhering to the gold standard systematic review methodology. Sensitivity for the hybrid human crowdsourcing and ML was 98%.
CONCLUSIONS: A collaborative, global PICU team integrated with ML was successful in efficient and accurate large data synthesis, producing a scoping open-access database of studies reporting on post-PICU outcomes. The development of this repository has implications for future reviews, providing opportunities for networking and collaborative engagement in research. The next steps should examine database maintenance, utilization, and dissemination of research findings.
DATA SOURCES: We performed a registered scoping review (OSF DOI10.17605/OSF.IO/HE5VB; Registered November 21, 2022) using MEDLINE, Embase, CINAHL, and CENTRAL databases, 2000-2022, with no language restrictions.
STUDY SELECTION: Observational or interventional studies describing outcomes of children (0-17 yr old) and their families or caregivers measured greater than 2 weeks post-PICU discharge. Titles and abstracts and full texts were initially screened by a large team of PICU healthcare workers and researchers who were recruited as part of an Evidence Hackathon event at the 2022 World Federation of Pediatric Intensive and Critical Care Societies conference. Initial screening results from 5000 citations were used to develop and validate an ML algorithm, after which a hybrid human crowdsourcing and ML approach was used to screen the remaining 11,055 studies.
DATA EXTRACTION: Not applicable.
DATA SYNTHESIS: Of 16,055 eligible citations, 1,301 met the criteria at full text for inclusion in the database. The screening was completed in just under 2 months while adhering to the gold standard systematic review methodology. Sensitivity for the hybrid human crowdsourcing and ML was 98%.
CONCLUSIONS: A collaborative, global PICU team integrated with ML was successful in efficient and accurate large data synthesis, producing a scoping open-access database of studies reporting on post-PICU outcomes. The development of this repository has implications for future reviews, providing opportunities for networking and collaborative engagement in research. The next steps should examine database maintenance, utilization, and dissemination of research findings.