Healthcare Record Exception Handling for Accurate Data Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing healthcare information technology systems face challenges in handling exceptions during the retrieval and processing of health records due to variations in data formats, encoding, and customization across different healthcare IT systems, leading to incomplete or inaccurate data extraction and processing.
Innovation Solution
A system and method that detects exceptions in health records at various levels, provides insights for configuration changes, and uses machine learning to handle these exceptions, allowing for deferred processing until appropriate changes are made, and includes graphical user interfaces for operator intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing interoperability solutions skip over elements with abnormal data formats or character set encoding, then processing speed is maintained, but measurement precision and manufacturing precision deteriorate due to loss of information
Solution Approach 1:
The patent segments the health record processing into multiple levels (record level, element level, data value level) and handles exceptions at each level independently. This allows the system to maintain processing speed by handling simple exceptions automatically while preserving precision by carefully analyzing complex exceptions at appropriate levels.
Solution Approach 2:
The system provides feedback to operators about detected exceptions and allows operators to review and correct extracted data. This feedback mechanism ensures measurement precision is maintained while keeping productivity high by automating routine corrections and only involving operators for complex cases.
2Measurement precision
If operators manually review and correct every exception, then measurement precision improves, but loss of time increases
Solution Approach 1:
The system performs self-service by automatically detecting, recording, and providing insights about exceptions. It automatically generates corrected records when possible and only requires operator intervention for complex cases, thereby maintaining high precision while minimizing time loss.
Solution Approach 2:
The system performs preliminary actions by pre-processing records to identify and flag exceptions before they reach the operator. This preliminary exception detection and classification allows operators to focus only on complex cases, reducing overall processing time while maintaining precision.
3Reliability
If the system records exceptions at multiple levels with context, then reliability improves, but device complexity increases
Solution Approach 1:
The patent implements a universal exception handling framework that operates across multiple levels (record, element, data value) and handles various types of exceptions (format errors, character encoding, missing values, etc.). This multi-functional approach improves reliability without proportionally increasing complexity by using a unified processing mechanism.
Solution Approach 2:
The system uses a nested structure where exceptions are recorded at multiple levels with contextual information about the record, element, and data value. This nesting allows comprehensive exception tracking and reliable handling while managing complexity through hierarchical organization of exception data.
4Reliability
If the system provides operators with actionable insights and configuration change opportunities, then reliability improves, but ease of operation deteriorates
Solution Approach 1:
The system provides targeted feedback to operators about specific exceptions and suggested corrections. This focused feedback approach improves reliability by ensuring accurate data handling while maintaining ease of operation by guiding operators through clear, actionable insights rather than overwhelming them with raw exception data.
Data Source
AI summary
Methods, systems, and apparatuses to improve the handling of exceptions during the retrieval and processing of health records from various data sources are provided. During the retrieval and processing of health records, exceptions to typical behavior are recorded with context at the data extraction protocol level, at the health record level and at the level of elements with the document. Accordingly, insights may be developed and configurations, rules, or coding changes, based on the detected exceptions may be proposed. In some instances, an operator may be notified about the exceptions such that the operator may act on the insight. In some instances, the processing of extracted records (documents, messages) may be deferred until the operator has made appropriate changes to configuration, rules, or code. In some instances, the system may supplement and/or replace the operator with machine learning engines that act on the developed insights.