Skip to main content
Fig. 4 | BMC Medical Informatics and Decision Making

Fig. 4

From: Large-scale identification of social and behavioral determinants of health from clinical notes: comparison of Latent Semantic Indexing and Generative Pretrained Transformer (GPT) models

Fig. 4

Retrieval performance of LSI, GPT-3.5 or GPT-4 when combined with ICD-9 coding. Precision (upper panel), recall (middle panel) and F1 (lower panel) of ICD-9 combined with either LSI (orange lines), GPT-3.5 (cyan lines) and GPT-4 (blue lines). Values represent the mean (filled circle) and 95% confidence intervals (error bars) across the nine SBDH gold standard sets

Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us