Medical Text Structuring by Attribute-Switch Delimiter Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical text, such as interpretation reports, are unstructured and difficult to utilize for secondary purposes like statistical analysis due to their free-form nature, as highlighted by JP2016-151827A, which does not provide specific support for creating structured reports.
Innovation Solution
A medical information processing device that sets the switching of attributes in sentences as delimiters, performs structuration processes for each unit defined by these delimiters, and outputs the results before the medical text creation is complete, allowing for structured output of medical texts like interpretation reports.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If medical texts are created in free-form format, then ease of writing is improved, but ease of secondary analysis deteriorates
Solution Approach 1:
The patent segments the medical text into structured units based on delimiter detection. The text is divided into multiple segments according to predefined delimiters (such as organ names, examination types), allowing each segment to be independently processed and analyzed while maintaining the original free-form writing style.
Solution Approach 2:
The patent changes the parameter of text structure from unstructured to structured by detecting delimiters and organizing content accordingly. The same medical text can be represented in different structural parameters (hierarchical, sequential, categorical) without changing the original wording, enabling both easy writing and efficient analysis.
2Difficulty of detecting and measuring
If medical texts are fully structured, then ease of secondary analysis is improved, but device complexity increases
Solution Approach 1:
The patent performs preliminary structuration by detecting delimiters and organizing text segments before secondary analysis is needed. The text is pre-processed into a structured format with identified segments and delimiters, so that subsequent statistical analysis and content analysis can be performed efficiently without complex real-time processing.
Solution Approach 2:
The system uses the medical text's own characteristics (presence of specific keywords, delimiters, and patterns) to automatically structure itself. The text contains inherent structural cues (organ names, examination types, conclusion markers) that the system leverages to perform self-organization, reducing the need for external complex processing mechanisms.
3Ease of operation
If structuration is performed after text creation, then ease of writing is improved, but loss of time in processing increases
Solution Approach 1:
The patent enables continuous structuration processing that can operate concurrently with text creation or immediately upon text input. The delimiter detection and segment organization processes run continuously as text is entered or received, eliminating idle time between writing and structuring, and enabling real-time or near-real-time availability of structured data for analysis.
Data Source
AI summary
Provided are a medical information processing device, a method for operating a medical information processing device, and a program which can provide support for creating a new medical text.A sentence input by a user is acquired in response to the input. Switching of an attribute of information described in the sentence is determined as a delimiter of the sentence. A structuration process is performed for each unit delimited by the delimiter. The determination is ended according to an end of creation of a medical text including one or more of the sentences. A result of the structuration process is output.