Machine learning models can predict subsequent publication of North American Spine Society (NASS) annual general meeting abstracts.


Journal

PloS one
ISSN: 1932-6203
Titre abrégé: PLoS One
Pays: United States
ID NLM: 101285081

Informations de publication

Date de publication:
2023
Historique:
received: 08 07 2022
accepted: 27 07 2023
medline: 22 8 2023
pubmed: 22 8 2023
entrez: 22 8 2023
Statut: epublish

Résumé

Academic meetings serve as an opportunity to present and discuss novel ideas. Previous studies have identified factors predictive of publication without generating predictive models. Machine learning (ML) presents a novel tool capable of generating these models. As such, the objective of this study was to use ML models to predict subsequent publication of abstracts presented at a major surgical conference. Database study. All abstracts from the North American Spine Society (NASS) annual general meetings (AGM) from 2013-2015 were reviewed. The following information was extracted: number of authors, institution, location, conference category, subject category, study type, data collection methodology, human subject research, and FDA approval. Abstracts were then searched on the PubMed, Google Scholar, and Scopus databases for publication. ML models were trained to predict whether the abstract would be published or not. Quality of models was determined by using the area under the receiver operator curve (AUC). The top ten most important factors were extracted from the most successful model during testing. A total of 1119 abstracts were presented, with 553 (49%) abstracts published. During training, the model with the highest AUC and accuracy metrics was the partial least squares (AUC of 0.77±0.05, accuracy of 75.5%±4.7%). During testing, the model with the highest AUC and accuracy was the random forest (AUC of 0.69, accuracy of 67%). The top ten features for the random forest model were (descending order): number of authors, year, conference category, subject category, human subjects research, continent, and data collection methodology. This was the first study attempting to use ML to predict the publication of complete articles after abstract presentation at a major academic conference. Future studies should incorporate deep learning frameworks, cognitive/results-based variables and aim to apply this methodology to larger conferences across other fields of medicine to improve the quality of works presented.

Sections du résumé

BACKGROUND CONTEXT BACKGROUND
Academic meetings serve as an opportunity to present and discuss novel ideas. Previous studies have identified factors predictive of publication without generating predictive models. Machine learning (ML) presents a novel tool capable of generating these models. As such, the objective of this study was to use ML models to predict subsequent publication of abstracts presented at a major surgical conference.
STUDY DESIGN/SETTING METHODS
Database study.
METHODS METHODS
All abstracts from the North American Spine Society (NASS) annual general meetings (AGM) from 2013-2015 were reviewed. The following information was extracted: number of authors, institution, location, conference category, subject category, study type, data collection methodology, human subject research, and FDA approval. Abstracts were then searched on the PubMed, Google Scholar, and Scopus databases for publication. ML models were trained to predict whether the abstract would be published or not. Quality of models was determined by using the area under the receiver operator curve (AUC). The top ten most important factors were extracted from the most successful model during testing.
RESULTS RESULTS
A total of 1119 abstracts were presented, with 553 (49%) abstracts published. During training, the model with the highest AUC and accuracy metrics was the partial least squares (AUC of 0.77±0.05, accuracy of 75.5%±4.7%). During testing, the model with the highest AUC and accuracy was the random forest (AUC of 0.69, accuracy of 67%). The top ten features for the random forest model were (descending order): number of authors, year, conference category, subject category, human subjects research, continent, and data collection methodology.
CONCLUSIONS CONCLUSIONS
This was the first study attempting to use ML to predict the publication of complete articles after abstract presentation at a major academic conference. Future studies should incorporate deep learning frameworks, cognitive/results-based variables and aim to apply this methodology to larger conferences across other fields of medicine to improve the quality of works presented.

Identifiants

pubmed: 37607198
doi: 10.1371/journal.pone.0289931
pii: PONE-D-22-18942
pmc: PMC10443859
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Pagination

e0289931

Informations de copyright

Copyright: © 2023 Abbas et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Déclaration de conflit d'intérêts

The authors have declared that no competing interests exist.

Références

J Foot Ankle Surg. 2019 May;58(3):410-416
pubmed: 30803914
J Thorac Oncol. 2010 Sep;5(9):1315-6
pubmed: 20736804
J Surg Orthop Adv. 2016 Summer;25(2):86-8
pubmed: 27518291
Global Spine J. 2018 May;8(3):273-278
pubmed: 29796376
J Arthroplasty. 2017 Nov 9;:1247-1252.e1
pubmed: 29174763
Eur Spine J. 2012 Oct;21(10):2105-12
pubmed: 22398641
Spine (Phila Pa 1976). 2017 Nov 15;42(22):1723-1729
pubmed: 28422799
Lancet Oncol. 2019 May;20(5):e262-e273
pubmed: 31044724
J Bone Joint Surg Am. 2002 Apr;84(4):615-21
pubmed: 11940624
Am J Surg. 2016 Jan;211(1):166-71
pubmed: 26349584
Int J Spine Surg. 2018 Dec 21;12(6):713-717
pubmed: 30619675
Spine (Phila Pa 1976). 2018 Mar 5;43(19):1347-1354
pubmed: 29509653
Injury. 2007 Jul;38(7):745-9
pubmed: 16978627
JAMA Netw Open. 2019 Jan 4;2(1):e186937
pubmed: 30646206
J Adv Pract Oncol. 2012 Mar;3(2):117-22
pubmed: 25059293
Arthroscopy. 2018 Mar;34(3):884-888
pubmed: 29249588

Auteurs

Aazad Abbas (A)

Division of Orthopaedic Surgery, University of Toronto, Toronto, ON, Canada.

Olumide Olotu (O)

Division of Orthopaedic Surgery, University of Western Ontario, London, ON, Canada.

Akeshdeep Bhatia (A)

Division of Orthopaedic Surgery, University of Toronto, Toronto, ON, Canada.

Denis Selimovic (D)

School of Medicine, St. George's University, University Centre, Grenada, West Indies.

Alireza Tajik (A)

School of Medicine, St. George's University, University Centre, Grenada, West Indies.

Jeremie Larouche (J)

Division of Orthopaedic Surgery, University of Toronto, Toronto, ON, Canada.
Division of Orthopaedic Surgery, Sunnybrook Health Science Centre, Toronto, ON, Canada.

Henry Ahn (H)

Division of Orthopaedic Surgery, University of Toronto, Toronto, ON, Canada.
Division of Orthopaedic Surgery, St. Michael's Hospital, Toronto, ON, Canada.

Albert Yee (A)

Division of Orthopaedic Surgery, University of Toronto, Toronto, ON, Canada.
Division of Orthopaedic Surgery, Sunnybrook Health Science Centre, Toronto, ON, Canada.

Stephen Lewis (S)

Division of Orthopaedic Surgery, University of Toronto, Toronto, ON, Canada.
Division of Orthopaedic Surgery Toronto Western Hospital, Toronto, ON, Canada.

Joel Finkelstein (J)

Division of Orthopaedic Surgery, University of Toronto, Toronto, ON, Canada.
Division of Orthopaedic Surgery, Sunnybrook Health Science Centre, Toronto, ON, Canada.

Jay Toor (J)

Department of Orthopaedic Surgery, University of Manitoba, Winnipeg, MB, Canada.

Classifications MeSH