From missing mPAP to clinical implementation: does automated imputation reveal missed pulmonary hypertension?

Extract Shams et al. [1] describe the development of an agentic large language model combined with machine learning capable of extracting pulmonary artery pressure values from unstructured right heart catheterisation reports, recovering information that was previously unavailable for analysis [2]. This study demonstrates the potential of artificial intelligence to transform unstructured clinical data into meaningful information for research and clinical care.