EUVIMEDOpdag EUVIMED
Open access

Use of Large Language Models to Assess the Likelihood of Epidemics From the Content of Tweets: Infodemiology Study

Michael S Deiner iD, Natalie A Deiner iD, Vagelis Hristidis iD, Stephen D McLeod iD, Thuy Doan iD, Thomas M Lietman iD, Travis C Porco iD

Journal of Medical Internet Research · 2024-03-01 · 26 · e49139

DOI: 10.2196/49139 ↗

At a glance

Journal article in Journal of Medical Internet Research, published online on 2024-03-01. Written by Michael S Deiner and 6 co-authors. Institution: Department of Ophthalmology, University of California, San Francisco, San Francisco, CA, United States. Major MEDLINE topics: Epidemics, Conjunctivitis, Social Media. Cited 36 times so far (Crossref); 50 references. The full text is freely available in PubMed Central. Funded by: NEI NIH HHS.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Background

Previous work suggests that Google searches could be useful in identifying conjunctivitis epidemics. Content-based assessment of social media content may provide additional value in serving as early indicators of conjunctivitis and other systemic infectious diseases.

Objective

We investigated whether large language models, specifically GPT-3.5 and GPT-4 (OpenAI), can provide probabilistic assessments of whether social media posts about conjunctivitis could indicate a regional outbreak.

Methods

A total of 12,194 conjunctivitis-related tweets were obtained using a targeted Boolean search in multiple languages from India, Guam (United States), Martinique (France), the Philippines, American Samoa (United States), Fiji, Costa Rica, Haiti, and the Bahamas, covering the time frame from January 1, 2012, to March 13, 2023. By providing these tweets via prompts to GPT-3.5 and GPT-4, we obtained probabilistic assessments that were validated by 2 human raters. We then calculated Pearson correlations of these time series with tweet volume and the occurrence of known outbreaks in these 9 locations, with time series bootstrap used to compute CIs.

Results

Probabilistic assessments derived from GPT-3.5 showed correlations of 0.60 (95% CI 0.47-0.70) and 0.53 (95% CI 0.40-0.65) with the 2 human raters, with higher results for GPT-4. The weekly averages of GPT-3.5 probabilities showed substantial correlations with weekly tweet volume for 44% (4/9) of the countries, with correlations ranging from 0.10 (95% CI 0.0-0.29) to 0.53 (95% CI 0.39-0.89), with larger correlations for GPT-4. More modest correlations were found for correlation with known epidemics, with substantial correlation only in American Samoa (0.40, 95% CI 0.16-0.81).

Conclusions

These findings suggest that GPT prompting can efficiently assess the content of social media posts and indicate possible disease outbreaks to a degree of accuracy comparable to that of humans. Furthermore, we found that automated content analysis of tweets is related to tweet volume for conjunctivitis-related posts in some locations and to the occurrence of actual epidemics. Future work may improve the sensitivity and specificity of these methods for disease outbreak detection.

Subjects

MeSH terms (MEDLINE)

EpidemicsConjunctivitisSocial MediaHumansUnited StatesInfodemiologyDisease OutbreaksLanguage

Author keywords

GPT-3.5GPT-4Generative Pre-trained TransformersTwitterX formerly known as Twitterconjunctivitisepidemic detectiongenerative large language modelinfectious eye diseasemicroblogsocial media

Authors and institutions

  • Department of Ophthalmology, University of California, San Francisco, San Francisco, CA, United States.
  • Francis I. Proctor Foundation for Research in Ophthalmology, University of California, San Francisco, San Francisco, CA, United States.
  • College of Letters and Science, University of California, Santa Barbara, Santa Barbara, CA, United States.
  • Department of Computer Science and Engineering, University of California, Riverside, Riverside, CA, United States.
  • American Academy of Ophthalmology, San Francisco, CA, United States.
  • Department of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, CA, United States.

ORCID profiles: Michael S Deiner, Natalie A Deiner, Vagelis Hristidis, Stephen D McLeod, Thuy Doan, Thomas M Lietman, Travis C Porco

Funding

  • NEI NIH HHS

Cite

APA

Deiner, M. S., Deiner, N. A., Hristidis, V., McLeod, S. D., Doan, T., Lietman, T. M., & Porco, T. C. (2024). Use of Large Language Models to Assess the Likelihood of Epidemics From the Content of Tweets: Infodemiology Study. Journal of Medical Internet Research, 26, e49139. https://doi.org/10.2196/49139

Vancouver

Deiner MS, Deiner NA, Hristidis V, McLeod SD, Doan T, Lietman TM, et al. Use of Large Language Models to Assess the Likelihood of Epidemics From the Content of Tweets: Infodemiology Study. Journal of Medical Internet Research. 2024;26:e49139. doi:10.2196/49139

BibTeX
@article{Deiner2024,
  author = {Deiner, Michael S and Deiner, Natalie A and Hristidis, Vagelis and McLeod, Stephen D and Doan, Thuy and Lietman, Thomas M and Porco, Travis C},
  title = {Use of Large Language Models to Assess the Likelihood of Epidemics From the Content of Tweets: Infodemiology Study},
  journal = {Journal of Medical Internet Research},
  year = {2024},
  volume = {26},
  pages = {e49139},
  doi = {10.2196/49139}
}
Start for free →