Clinical Digital Assistant Pipeline for EHR Entity Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Clinical environments face inefficiencies and errors due to reliance on multiple devices and trained personnel for documenting patient encounters and managing electronic health records, leading to cumbersome, resource-intensive, and costly healthcare processes.
Innovation Solution
A clinical digital assistant processing pipeline that uses two machine-learning models for medical entity detection and resolution, generating a FHIR-compliant data structure to accurately and efficiently populate electronic health records and facilitate medication orders.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple devices and trained personnel are used for documenting patient encounters and managing electronic health records, then the quality and accuracy of healthcare documentation can be maintained, but the process becomes cumbersome, time-intensive, and costly
Solution Approach 1:
The system enables automatic documentation by having the digital assistant independently perform entity detection, classification, and code mapping without requiring manual intervention from healthcare providers. The machine learning models automatically process clinical notes and generate structured EHR data, allowing the system to serve itself rather than requiring trained personnel for each documentation task.
Solution Approach 2:
The patent replaces the mechanical process of manual documentation with an automated digital assistant system using machine learning models. The entity detection model, classification model, and code mapping model collectively substitute the manual work of healthcare providers, transforming the mechanical documentation process into an automated computational system that reduces time and resource requirements while maintaining documentation quality.
2Reliability
If multiple devices and trained personnel are used for documenting patient encounters, then accurate patient records can be created, but the process becomes resource-intensive and costly
Solution Approach 1:
The digital assistant system performs multiple functions including entity detection, classification, code mapping, and EHR population through a single integrated platform. The machine learning models work together to handle various documentation tasks that previously required multiple different devices and personnel types, consolidating resources while maintaining record accuracy.
Solution Approach 2:
The system automatically detects medical entities, classifies them according to FHIR standards, maps them to appropriate medical codes, and populates EHR fields without requiring external resources. This self-service capability eliminates the need for multiple trained personnel and devices while maintaining the accuracy previously achieved through manual processes.
3Adaptability or versatility
If traditional manual methods are used for healthcare documentation, then flexibility in handling diverse clinical scenarios is maintained, but efficiency and productivity are reduced
Solution Approach 1:
The machine learning models are designed to dynamically adapt to different clinical scenarios through their training on diverse medical data. The entity detection and classification models can handle various types of clinical notes, patient encounters, and medical terminology, providing flexibility comparable to manual methods while achieving much higher processing speeds and productivity.
Data Source
AI summary
The techniques described herein provide a novel clinical digital assistant (CDA) processing pipeline enabling medical entity detection and resolution that works against various EHRs and with different ontologies (e.g., medical coding systems). In some embodiments, the processing pipeline may involve two machine-learning models that can perform named entity recognition on the natural language utterance to identify medical entities that are associated with different medical entity types, and link the medical entities to medical codes of standard medical coding systems. A FHIR-compliance data structure may be generated using the identified medical codes, their associated medical coding systems, the identified medical entities, and their associated medical entity types.