Evaluating ChatGPT text mining of clinical records for companion animal obesity monitoring.

Abstract

BACKGROUND: Veterinary clinical narratives remain a largely untapped resource for addressing complex diseases. Here we compare the ability of a large language model (ChatGPT) and a previously developed regular expression (RegexT) to identify overweight body condition scores (BCS) in veterinary narratives pertaining to companion animals. METHODS: BCS values were extracted from 4415 anonymised clinical narratives using either RegexT or by appending the narrative to a prompt sent to ChatGPT, prompting the model to return the BCS information. Data were manually reviewed for comparison. RESULTS: The precision of RegexT was higher (100%, 95% confidence interval [CI] 94.81%-100%) than that of ChatGPT (89.3%, 95% CI 82.75%-93.64%). However, the recall of ChatGPT (100%, 95% CI 96.18%-100%) was considerably higher than that of RegexT (72.6%, 95% CI 63.92%-79.94%). LIMITATIONS: Prior anonymisation and subtle prompt engineering are needed to improve ChatGPT output. CONCLUSIONS: Large language models create diverse opportunities and, while complex, present an intuitive interface to information. However, they require careful implementation to avoid unpredictable errors.