Deep reflective reasoning in interdependence constrained structured data extraction from clinical notes for digital health
This addresses the challenge of ensuring clinically consistent data extraction from notes for digital health applications, representing a domain-specific incremental improvement over existing LLM pipelines.
The paper tackled the problem of extracting structured information from clinical notes with interdependent variables, where existing LLM-based methods often produce inconsistent outputs, and proposed a deep reflective reasoning framework that iteratively self-critiques and revises outputs to improve consistency, resulting in significant accuracy gains such as increasing average F1 from 0.828 to 0.911 in colorectal cancer reporting and improving tumor stage accuracy from 0.680 to 0.833 in lung cancer.
Extracting structured information from clinical notes requires navigating a dense web of interdependent variables where the value of one attribute logically constrains others. Existing Large Language Model (LLM)-based extraction pipelines often struggle to capture these dependencies, leading to clinically inconsistent outputs. We propose deep reflective reasoning, a large language model agent framework that iteratively self-critiques and revises structured outputs by checking consistency among variables, the input text, and retrieved domain knowledge, stopping when outputs converge. We extensively evaluate the proposed method in three diverse oncology applications: (1) On colorectal cancer synoptic reporting from gross descriptions (n=217), reflective reasoning improved average F1 across eight categorical synoptic variables from 0.828 to 0.911 and increased mean correct rate across four numeric variables from 0.806 to 0.895; (2) On Ewing sarcoma CD99 immunostaining pattern identification (n=200), the accuracy improved from 0.870 to 0.927; (3) On lung cancer tumor staging (n=100), tumor stage accuracy improved from 0.680 to 0.833 (pT: 0.842 -> 0.884; pN: 0.885 -> 0.948). The results demonstrate that deep reflective reasoning can systematically improve the reliability of LLM-based structured data extraction under interdependence constraints, enabling more consistent machine-operable clinical datasets and facilitating knowledge discovery with machine learning and data science towards digital health.