A recently completed study at the University Hospital of Toulouse, France emphasized the benefits of a "domain-specific natural language processing (NLP) pipeline...in extracting clinically meaningful information from diverse clinical documents of patients with non-small cell lung cancer." Following an analysis of 1,028 discharge summaries and external consultation letters from 120 non-small cell lung cancer patients undergoing oral targeted therapy, "the domain-adapted NLP solution achieved an F1 score of 79.7% for tumor evolution concept extraction and 62.0% for temporality alignment."
To read more about this study, click here.
Source mentioned:
Vinot, C., et al. "Automated extraction of temporalized tumor evolution from oncology EMRs using natural language processing." ESMO Real World Data and Digital Oncology 11 (2026): 100660.