Clinical Information System Semantic Search Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Clinical users face challenges in efficiently finding key information from large volumes of medical data due to the difficulty in understanding the current clinical context, leading to information overload and inefficient use of automated systems.
Innovation Solution
A clinical information system that uses a text-expanding semantic search function to identify related terms in unstructured clinical notes, and rules circuitry to execute associated algorithms on structured data, providing notifications to the user about relevant findings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated systems run all available tasks on all data when an Electronic Medical Record is opened, then all potential findings are identified, but hardware costs and processing time increase significantly
Solution Approach 1:
The system performs preliminary actions by analyzing user input and search behavior before executing full automated tasks. It predicts which tasks the user likely wants to perform and pre-processes or prioritizes those specific tasks, avoiding the need to run all available tasks on all data while still identifying all relevant findings.
Solution Approach 2:
The system changes the parameter of task execution from 'all tasks on all data' to 'predicted tasks on relevant data'. It dynamically adjusts which automated tasks are run based on user input analysis, transforming the exhaustive approach into a targeted, efficient approach that maintains finding completeness while reducing hardware resource usage.
2Loss of information
If automated systems analyze all data to provide comprehensive information, then all relevant findings are presented, but information overload occurs and key information becomes difficult to find
Solution Approach 1:
The system extracts only the key information and findings that are most relevant to the user's current clinical context and search intent. Instead of presenting all available findings, it selectively extracts and highlights the most important ones, maintaining information completeness for decision-making while improving ease of operation by reducing visual clutter and distraction.
Solution Approach 2:
The system applies local quality by providing different levels of information presentation based on user needs and context. Key findings are highlighted with higher visibility and priority, while less critical information is presented with lower prominence. This creates a differentiated information hierarchy that maintains completeness while improving retrieval efficiency through strategic emphasis.
3Loss of information
If clinicians manually search for key information in large volumes of medical data, then they can find relevant information, but time consumption and workload increase
Solution Approach 1:
The system performs self-service by automatically analyzing user input, predicting intended tasks, and executing relevant automated tasks without requiring manual intervention. It autonomously identifies and presents key findings, eliminating the need for clinicians to manually search through large volumes of data while maintaining accurate information retrieval, thus reducing both time consumption and workload.
Solution Approach 2:
The system uses feedback from user input and search behavior to continuously improve its predictions and task selections. By monitoring what users search for and how they interact with the system, it refines its ability to anticipate user needs and automatically retrieve the correct information, reducing time loss while maintaining finding accuracy through adaptive learning.
4Loss of information
If text-expanding semantic search is used to find related terms in clinical notes, then comprehensive search results are provided, but processing complexity increases
Solution Approach 1:
The system performs preliminary action by generating related terms and expanding the search query before executing the actual search on clinical notes. It pre-processes the search input to include semantically related terms, which simplifies the subsequent search execution while maintaining comprehensive search results. This preliminary term expansion reduces the complexity of the main search operation by preparing the query in advance.
Solution Approach 2:
The text-expanding semantic search function acts as an intermediary between the user's simple search input and the complex task of finding all relevant information in clinical notes. It transforms the user's basic search term into an expanded set of related terms, mediating the complexity by handling the term generation and expansion process separately, which simplifies the overall search system architecture while maintaining search completeness.
Data Source
AI summary
A clinical information system comprises processing circuitry configured to: receive a user input from a user, wherein the user input instructs the performing of a first action on first medical data for a subject; determine based on the user input and/or the first action at least one input term; determine at least one further term that is conceptually related to the at least one input term; determine whether any stored action of a set of stored actions is associated with the at least one further term; and if a stored action is associated with the at least one further term: perform said stored action on second medical data for the subject; and provide to the user a notification of said stored action.