Automated Medical Problem List Generation from EMR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic medical records are challenging to analyze due to their size and complexity, often lacking an accurate list of clinical concerns relevant to patient care, which requires significant medical expertise and is typically incomplete or overly inclusive when generated by conventional systems.
Innovation Solution
The system maps both structured and unstructured data from electronic medical records to standardized medical concepts using natural language processing, applying techniques like information extraction, text segmentation, and latent semantic analysis to generate feature values and weight medical problem concepts, identifying relevant issues with a weighted scoring system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional systems are used to generate problem lists, then the system is simple to operate, but the accuracy and reliability of the medical problem list is poor
Solution Approach 1:
The patent introduces multiple intermediary components including a normalization module that maps medical terms to standardized concepts, a feature extraction module that identifies clinical indicators, and a weighting module that applies clinical expertise weights. These intermediaries transform raw unstructured EMR data into structured, weighted problem lists, resolving the contradiction between accuracy and complexity by inserting specialized processing layers.
Solution Approach 2:
The patent replaces manual medical expertise review (mechanical human process) with an automated computational system that uses natural language processing, statistical weighting, and algorithmic ranking. This substitution maintains high accuracy while reducing operational complexity for end users, as the system automatically performs tasks that previously required expert clinicians.
2Measurement precision
If manual expert review is used to create problem lists, then the accuracy is high, but the time consumption and productivity are low
Solution Approach 1:
The system enables self-service by automatically generating problem lists without requiring manual expert review. The automated pipeline includes document parsing, entity recognition, feature extraction, and weighted scoring that operates independently, producing clinically accurate problem lists at scale without human intervention in the generation process.
Solution Approach 2:
The patent performs preliminary processing of EMR data by pre-normalizing medical terms to standardized concepts, pre-extracting relevant clinical features, and pre-calculating weights based on clinical importance. This preliminary action prepares data in advance, enabling rapid final problem list generation while maintaining accuracy that would otherwise require time-consuming manual review.
3Loss of information
If comprehensive analysis of all EMR data is performed, then the completeness of medical problems is improved, but the computational complexity and processing time increase
Solution Approach 1:
The patent extracts only the most relevant clinical features and indicators from the comprehensive EMR data using targeted pattern matching and feature selection algorithms. Instead of analyzing all data equally, the system identifies and extracts specific clinical indicators, lab results, and medication patterns that are most predictive of actual medical problems, reducing processing complexity while maintaining completeness.
Solution Approach 2:
The patent applies different processing qualities to different parts of the EMR data based on clinical importance. High-weight features receive more rigorous analysis and validation, while lower-weight features receive streamlined processing. This local quality approach ensures comprehensive coverage of important medical problems while avoiding unnecessary computational complexity in less critical areas.
Data Source
AI summary
Methods, systems, and devices map data of an electronic medical record to standardized medical concepts. The standardized medical concepts are defined by medical industry standards organizations. The methods, systems, and devices identify medical problem concepts, from the standardized medical concepts, generate feature values of the medical problem concepts based on features within the mapped standardized medical concepts, and weight the medical problem concepts based on the feature values according to a weighting function. These methods, systems, and devices identify medical problems as ones of the medical problem concepts that have a weighted score above a threshold, according to the weighting, and output a list of the medical problems.