Hiring in life sciences? Share your open positions with our professional community. Read more Close

Advertisement

Unsupervised topic exploration of the "Brain Disease Patient Community": A text mining approach using latent Dirichlet allocation.

Created on 15 Nov 2026

Authors

Yu Seong Hwang, Yejun Kim, Jin Su Shin, Heui Sug Jo, Hocheol Lee

Published in

Digital health. Volume 12. Pages 20552076261493326. Epub Sep 24, 2026.

Abstract

Online patient communities accumulate a large volume of user-generated narratives regarding symptoms, treatment decisions, rehabilitation, caregiving, and administrative processes. Systematically characterizing these narratives may provide patient-centered insights that complement conventional clinical and survey-based research. This study aimed to identify and interpret major themes discussed in a South Korean online community for patients with brain diseases, with a focus on ischemic stroke (cerebral infarction), spontaneous intracerebral hemorrhage, and cerebral aneurysm.
A retrospective cross-sectional text-mining study was conducted on posts and associated comments published from October 1, 2024, to March 31, 2025, in a large Naver Cafe community. Posts were archived as PDF files, converted into text, and merged into a single corpus. Korean-language preprocessing was performed using the KoNLPy package with the MeCab morphological analyzer. Latent Dirichlet allocation (LDA) topic models were fitted across candidate numbers of topics. The final model was selected using quantitative metrics (including coherence and perplexity) and interpretability. Topic interpretation and labeling were supported by pyLDAvis keyword relevance weighting (λ=0.6) and expert review.
After preprocessing, the final corpus comprised 5,298 posts (including comments) and 20,474 unique tokens. The optimal LDA solution identified 12 topics clustered into four broad domains: (i) clinical management and secondary prevention, (ii) recovery and rehabilitation, (iii) inpatient care and caregiving, and (iv) psychosocial and administrative needs.
Unsupervised topic modeling of online community discussions delineated a coherent set of patient- and caregiver-expressed needs that extend beyond clinical recovery to include caregiving burden and navigation of administrative and financial support systems.

PMID:
43034442
Bibliographic data and abstract were imported from PubMed on 15 Nov 2026.

Read full publication at:
Please sign in to see all details.

Advertisement

Stats

  • Community rating n/a 0 votes
  • Reviewers' rating n/a 0 votes
  • Your rating

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

  • Recommendations n/a n/a positive of 0 vote(s)
  • Views 12
  • Comments 0

Recommended by

  • No recommendations yet.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.

Advertisement