PMID- 38512919
OWN - NLM
STAT- MEDLINE
DCOM- 20240325
LR  - 20240325
IS  - 1932-6203 (Electronic)
IS  - 1932-6203 (Linking)
VI  - 19
IP  - 3
DP  - 2024
TI  - AE-GPT: Using Large Language Models to extract adverse events from surveillance 
      reports-A use case with influenza vaccine adverse events.
PG  - e0300919
LID - 10.1371/journal.pone.0300919 [doi]
LID - e0300919
AB  - Though Vaccines are instrumental in global health, mitigating infectious diseases 
      and pandemic outbreaks, they can occasionally lead to adverse events (AEs). 
      Recently, Large Language Models (LLMs) have shown promise in effectively 
      identifying and cataloging AEs within clinical reports. Utilizing data from the 
      Vaccine Adverse Event Reporting System (VAERS) from 1990 to 2016, this study 
      particularly focuses on AEs to evaluate LLMs' capability for AE extraction. A 
      variety of prevalent LLMs, including GPT-2, GPT-3 variants, GPT-4, and Llama2, 
      were evaluated using Influenza vaccine as a use case. The fine-tuned GPT 3.5 
      model (AE-GPT) stood out with a 0.704 averaged micro F1 score for strict match 
      and 0.816 for relaxed match. The encouraging performance of the AE-GPT 
      underscores LLMs' potential in processing medical data, indicating a significant 
      stride towards advanced AE detection, thus presumably generalizable to other AE 
      extraction tasks.
CI  - Copyright: (c) 2024 Li 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.
FAU - Li, Yiming
AU  - Li Y
AUID- ORCID: 0009-0009-8784-1745
AD  - McWilliams School of Biomedical Informatics, The University of Texas Health 
      Science Center at Houston, Houston, TX, United States of America.
FAU - Li, Jianfu
AU  - Li J
AD  - Department of Artificial Intelligence and Informatics, Mayo Clinic, Jacksonville, 
      FL, United States of America.
FAU - He, Jianping
AU  - He J
AD  - McWilliams School of Biomedical Informatics, The University of Texas Health 
      Science Center at Houston, Houston, TX, United States of America.
FAU - Tao, Cui
AU  - Tao C
AUID- ORCID: 0000-0002-4267-1924
AD  - Department of Artificial Intelligence and Informatics, Mayo Clinic, Jacksonville, 
      FL, United States of America.
LA  - eng
PT  - Journal Article
DEP - 20240321
PL  - United States
TA  - PLoS One
JT  - PloS one
JID - 101285081
RN  - 0 (Influenza Vaccines)
RN  - EC 2.6.1.2 (Alanine Transaminase)
SB  - IM
MH  - Humans
MH  - *Influenza Vaccines/adverse effects
MH  - Adverse Drug Reaction Reporting Systems
MH  - *Influenza, Human/prevention & control
MH  - Alanine Transaminase
MH  - Disease Outbreaks
PMC - PMC10956752
COIS- The authors have declared that no competing interests exist.
EDAT- 2024/03/21 18:45
MHDA- 2024/03/25 06:43
PMCR- 2024/03/21
CRDT- 2024/03/21 13:53
PHST- 2023/10/12 00:00 [received]
PHST- 2024/03/06 00:00 [accepted]
PHST- 2024/03/25 06:43 [medline]
PHST- 2024/03/21 18:45 [pubmed]
PHST- 2024/03/21 13:53 [entrez]
PHST- 2024/03/21 00:00 [pmc-release]
AID - PONE-D-23-33399 [pii]
AID - 10.1371/journal.pone.0300919 [doi]
PST - epublish
SO  - PLoS One. 2024 Mar 21;19(3):e0300919. doi: 10.1371/journal.pone.0300919. 
      eCollection 2024.