AI and RAG Flowsheet Population for Speech-Based Documentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manually determining relevant rows in hospital-specific flowsheets for patient data documentation is time-consuming and inefficient, leading to increased cognitive burden, delayed accessibility, and potential inaccuracies in healthcare documentation.
Innovation Solution
A system utilizing generative artificial intelligence (AI) and retrieval augmented generation (RAG) to automatically populate relevant rows in flowsheets by processing speech inputs, segmenting meaningful information, and applying similarity-based searches to match transcript segments with appropriate schema and example databases, ensuring accurate and efficient data entry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual determination of relevant rows is used, then data accuracy can be maintained through human judgment, but nursing personnel efficiency deteriorates due to time-consuming manual work
Solution Approach 1:
The system enables self-service by automatically populating flowsheet rows using speech-to-text conversion of nurse dictation. The natural language processing system independently identifies relevant parameters and extracts values without requiring manual row selection or data entry, allowing nurses to document patient data through simple speech rather than manual interface interaction.
Solution Approach 2:
The patent replaces the mechanical manual process of row selection and data entry with an automated speech processing system. Speech-to-text conversion and natural language algorithms substitute for manual cognitive and physical tasks, transforming the workflow from manual interface manipulation to automated speech-based documentation.
2Speed
If manual row selection is performed, then cognitive burden is reduced through human decision-making, but documentation speed deteriorates due to sequential manual processes
Solution Approach 1:
The system performs self-service by automatically determining which rows are relevant and populating them based on speech input. The natural language processing system independently analyzes the dictation, identifies applicable clinical parameters, and fills corresponding flowsheet rows without requiring nurse intervention for row selection or data entry.
Solution Approach 2:
The system performs preliminary action by pre-processing speech input through speech-to-text conversion and natural language analysis before flowsheet population. This preliminary processing identifies relevant parameters and extracts values in advance, preparing the data for automatic row population and eliminating the need for sequential manual operations.
3Productivity
If automated speech processing is implemented, then productivity is improved through reduced manual effort, but system complexity increases due to AI and RAG components
Solution Approach 1:
The patent uses an intermediary natural language processing layer between speech input and flowsheet population. This intermediary system includes speech-to-text conversion, natural language analysis, and parameter identification components that translate spoken dictation into structured flowsheet data, bridging the gap between unstructured speech and structured electronic health record entry.
Solution Approach 2:
The system replaces manual mechanical processes with automated speech processing technology. Speech-to-text conversion systems, natural language processing algorithms, and automated row population mechanisms substitute for manual cognitive and physical tasks, handling the complexity of data extraction and mapping automatically.
4Measurement precision
If comprehensive schema is used, then measurement precision is improved through detailed parameter tracking, but ease of operation deteriorates due to overwhelming number of rows
Solution Approach 1:
The system performs self-service by automatically analyzing speech input and identifying which comprehensive schema parameters are relevant to the current patient situation. The natural language processing system independently determines parameter relevance and populates only the appropriate rows from the comprehensive schema, filtering out irrelevant parameters without requiring manual selection.
Solution Approach 2:
The system extracts only the relevant parameters from the comprehensive schema based on speech input analysis. Rather than presenting or requiring entry into all available rows, the system extracts and populates only the specific parameters mentioned or implied in the nurse's dictation, filtering the comprehensive schema to show only relevant fields.
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 plurality of key/value variations associated with an instance of information are identified. A plurality of key/value variation vectors are generated by embedding each key/value variation in a vector. The plurality of key/value variation vectors are combined into a combined vector representing the instance of information. A transcript is segmented into a segment, the segment corresponding to the instance of information. A transcript segment vector is generated by embedding the instance of information in the segment into a vector. A similarity between the transcript segment vector and the and the combined vector is determined. The instance is extracted from the transcript segment vector by processing a prompt with a generative artificial intelligence (AI) model using retrieval augmented generation (RAG). A value of a row corresponding to the key is populated with the instance of information.