Zero-shot interpretable phenotyping of postpartum hemorrhage using large language models.

Many areas of medicine would benefit from deeper, more accurate phenotyping, but there are limited approaches for phenotyping using clinical notes without substantial annotated data. Large language models (LLMs) have demonstrated immense potential to adapt to novel tasks with no additional training by specifying task-specific instructions. Here we report the performance of a publicly available LLM, Flan-T5, in phenotyping patients with postpartum hemorrhage (PPH) using discharge notes from electronic health records (n = 271,081). The language model achieves strong performance in extracting 24 granular concepts associated with PPH. Identifying these granular concepts accurately allows the development of interpretable, complex phenotypes and subtypes. The Flan-T5 model achieves high fidelity in phenotyping PPH (positive predictive value of 0.95), identifying 47% more patients with this complication compared to the current standard of using claims codes. This LLM pipeline can be used reliably for subtyping PPH and outperforms a claims-based approach on the three most common PPH subtypes associated with uterine atony, abnormal placentation, and obstetric trauma. The advantage of this approach to subtyping is its interpretability, as each concept contributing to the subtype determination can be evaluated. Moreover, as definitions may change over time due to new guidelines, using granular concepts to create complex phenotypes enables prompt and efficient updating of the algorithm. Using this language modelling approach enables rapid phenotyping without the need for any manually annotated training data across multiple clinical use cases.

Zero-shot Interpretable Phenotyping of Postpartum Hemorrhage Using Large Language Models.

Many areas of medicine would benefit from deeper, more accurate phenotyping, but there are limited approaches for phenotyping using clinical notes without substantial annotated data. Large language models (LLMs) have demonstrated immense potential to adapt to novel tasks with no additional training by specifying task-specific i nstructions. We investigated the per-formance of a publicly available LLM, Flan-T5, in phenotyping patients with postpartum hemorrhage (PPH) using discharge notes from electronic health records ( n =271,081). The language model achieved strong performance in extracting 24 granular concepts associated with PPH. Identifying these granular concepts accurately allowed the development of inter-pretable, complex phenotypes and subtypes. The Flan-T5 model achieved high fidelity in phenotyping PPH (positive predictive value of 0.95), identifying 47% more patients with this complication compared to the current standard of using claims codes. This LLM pipeline can be used reliably for subtyping PPH and outperformed a claims-based approach on the three most common PPH subtypes associated with uterine atony, abnormal placentation, and obstetric trauma. The advantage of this approach to subtyping is its interpretability, as each concept contributing to the subtype determination can be evaluated. Moreover, as definitions may change over time due to new guidelines, using granular concepts to create complex phenotypes enables prompt and efficient updating of the algorithm. Using this lan-guage modelling approach enables rapid phenotyping without the need for any manually annotated training data across multiple clinical use cases.

Early recognition of intraventricular hemorrhage in the setting of thrombocytosis in the emergency department.

Thrombocytosis is frequently encountered as an incidental laboratory finding since isolated thrombocytosis is often asymptomatic. Even though thrombocytosis is benign and self-limiting in most cases, it can at times result in thrombosis or hemorrhage. The most common type of thrombocytosis is reactive (secondary) thrombocytosis and can be due to infections, trauma, surgery, or occult malignancy. Since thrombocytosis is a known risk factor for thrombosis, it is commonly a concern for ischemic stroke and myocardial infarction. Much less common are hemorrhagic events associated with thrombocytosis. Studies have shown that when hemorrhage is present in patients with thrombocytosis, it is most often seen in the setting of chronic myelogenous leukemia and essential thrombocythemia. In essential thrombocythemia, the overall risk of bleeding and thrombosis is 0.33% per patient-year and 6.6% per patient-year, respectively. In the general population, the risk of bleeding and thrombosis is 0% and 1.2%, respectively. The present study is a case report of an 83-year-old man who presented to the emergency department with hypertension and headache, who was then found to have significant thrombocytosis (platelets >1,000×10(9)/L) and acute right intraventricular hemorrhage without any signs of neurological deficits, or evidence of vascular malformations or mass. We present this case report for review and discussion of some of the challenges and considerations associated with the management of such patients.