Text-based predictions of COVID-19 diagnosis from self-reported chemosensory descriptions.
Hongyang LiRichard C GerkinAlyssa J BakkeRaquel NorelGuillermo A CecchiChristophe LaudamielMasha Y NivKathrin OhlaJohn E HayesValentina ParmaPablo MeyerPublished in: Communications medicine (2023)
Our results show that the description of perceptual symptoms caused by a viral infection can be used to fine-tune an LLM model to correctly predict and interpret the diagnostic status of a subject. In the future, similar models may have utility for patient verbatims from online health portals or electronic health records.
Keyphrases
- electronic health record
- health information
- coronavirus disease
- sars cov
- clinical decision support
- public health
- healthcare
- case report
- social media
- adverse drug
- air pollution
- working memory
- mental health
- current status
- smoking cessation
- sleep quality
- respiratory syndrome coronavirus
- risk assessment
- human health
- finite element