Task Assignment System for Medical Document Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in accurately classifying medical conditions from electronic medical records (EMRs), particularly for under-documented diseases like insomnia, due to insufficient diagnosis codes and prescription data, leading to unreliable data for machine learning algorithms.
Innovation Solution
A computer-implemented method and system for assigning assessment tasks to qualified reviewers based on task criteria and reviewer credentials, which involves determining tasks, selecting appropriate reviewers, providing documents for review, receiving results, and storing them in EMRs, while considering expertise to ensure reliable classifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual chart review is used to classify ambiguous medical conditions, then classification accuracy is improved, but time consumption and operational complexity increase
Solution Approach 1:
The patent segments the manual review process into structured task components with specific criteria (e.g., presence/absence of condition, confidence levels). Reviewers evaluate documents against predefined task criteria rather than performing comprehensive subjective analysis, reducing time while maintaining accuracy for ambiguous cases.
Solution Approach 2:
The system introduces quantitative parameters (confidence levels, task criteria scores) to transform subjective physician judgments into measurable data. This parameterization enables efficient processing while preserving classification accuracy by capturing the essence of expert judgment in structured form.
2Measurement precision
If manual chart review is used to classify ambiguous medical conditions, then classification accuracy is improved, but system complexity increases
Solution Approach 1:
The patent creates a universal task assignment system that handles multiple document types, conditions, and reviewer expertise levels through a single standardized platform. The system manages reviewer selection, task assignment, criteria evaluation, and result aggregation through integrated multi-functional modules, reducing overall system complexity despite the sophisticated classification tasks.
Solution Approach 2:
The system introduces an intermediary task management layer between the raw medical documents and the final classifications. This intermediary structure standardizes the interaction between reviewers and documents through predefined task criteria and templates, simplifying the complexity of coordinating multiple reviewers while maintaining high classification accuracy.
3Reliability
If reviewer selection is based on comprehensive credential matching, then classification reliability is improved, but selection complexity and time increase
Solution Approach 1:
The patent applies local quality by matching reviewer credentials specifically to the requirements of each task type and document category rather than requiring all reviewers to have universal expertise. The system identifies the specific knowledge domain needed for each assessment task and selects reviewers with corresponding specialized credentials, improving reliability while reducing selection complexity.
Solution Approach 2:
The system transforms the reviewer selection process by converting comprehensive credential evaluation into specific parameter matching (e.g., required credentials, expertise areas, task compatibility). This parameterized approach automates the selection process and reduces complexity while maintaining high classification reliability through targeted expert matching.
Data Source
AI summary
A system and method for assigning assessment tasks includes determining tasks in need of completion, where the tasks include assessing a set of documents containing medical patient data and providing a judgment (e.g., classification or label) based on the contents of the document. A process also includes selecting one or more reviewers based on the one or more tasks and providing one or more documents from the set of documents to each of the selected one or more reviewers for completion of the one or more tasks. The process further includes receiving a result of the one or more tasks after completion by the selected one or more reviewers and storing the result in an electronic medical record database.