Generative AI Flowsheet Population from Spoken Clinical Text
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Nurses face inefficiencies in completing flowsheets in electronic health records due to the manual process of determining relevant rows and entering data, which increases cognitive burden and delays in data accessibility.
Innovation Solution
A system and method for flowsheet population that uses a generative artificial intelligence model to automatically populate structured documents by converting spoken text into structured data, utilizing retrieval augmented generation to match transcript segments with appropriate rows in the flowsheet schema.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual process is used to determine relevant rows and enter data in flowsheets, then nurses can complete documentation tasks, but the cognitive burden increases and data accessibility is delayed
Solution Approach 1:
The patent replaces the manual mechanical process of nurses reviewing and selecting flowsheet rows with an automated natural language processing system. The system automatically extracts relevant information from clinical notes and populates the appropriate flowsheet rows, eliminating the manual cognitive burden and time loss associated with manual data entry and retrieval.
2Reliability
If all rows in the flowsheet schema are displayed to nurses, then complete data collection is possible, but the complexity of the interface increases and nurses are overwhelmed
Solution Approach 1:
The patent extracts and displays only the relevant subset of flowsheet rows that correspond to information actually present in the clinical note. The natural language processing system identifies which data points are mentioned in the provider's note and automatically populates only those specific rows, filtering out irrelevant rows and reducing interface complexity while maintaining data collection completeness for relevant information.
3Measurement precision
If manual determination of relevant rows is performed, then data accuracy can be maintained through nurse judgment, but the time required to complete flowsheets increases
Solution Approach 1:
The patent replaces the manual judgment process with an automated natural language processing system that uses trained algorithms to identify and extract relevant clinical information. The system achieves high accuracy through sophisticated NLP techniques including entity recognition, relationship extraction, and context analysis, while simultaneously reducing completion time by eliminating manual review and selection steps.
Data Source
AI summary
A method, computer program product, and computing system for flowsheet population. A structured document conforming to a schema and having a plurality of rows, each row comprising a key and a value, is processed. A target key/value pair is identified within the structured document for a target instance of information. An in-context example of spoken text representative of the target instance of information corresponding to the target key/value pair is generated using a generative artificial intelligence (AI) model.