Dynamic Prescription Auditing via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current prescription auditing systems face challenges due to non-uniform medical resource distribution, leading to pharmacists lacking knowledge and inaccuracies in auditing prescriptions, with existing rules being either manually configured or statically extracted from pharmacological knowledge graphs, resulting in inefficiencies and inaccuracies.
Innovation Solution
A method and apparatus that acquire prescription comment information, perform feature extraction, and use a classification decision model trained with machine learning to generate an updated auditing rule, dynamically learning from diverse hospital data to improve auditing accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If auditing rules are manually configured by pharmacists based on experience, then the rules can be updated flexibly, but the system complexity and manual workload increase
Solution Approach 1:
The patent transforms static manually-configured auditing rules into dynamic rules that are automatically updated through machine learning. The system continuously learns from new prescription data and automatically adjusts auditing rules without requiring manual reconfiguration, thereby maintaining flexibility while reducing system complexity and manual workload.
Solution Approach 2:
The patent implements a self-updating mechanism where the auditing system automatically improves its own rules through machine learning. The system independently processes new prescription data, extracts patterns, and updates auditing rules without human intervention, enabling the system to serve itself in maintaining and improving its auditing capabilities.
2Stability of the object's composition
If auditing rules are extracted from pharmacological knowledge graphs, then the rules can be systematically organized, but the rules become static and less adaptable to new situations
Solution Approach 1:
The patent transforms static rules extracted from knowledge graphs into dynamic rules that automatically adapt to new situations. By incorporating machine learning that processes new prescription data in real-time, the system maintains the systematic organization of knowledge graph-based rules while enabling continuous adaptation to emerging patterns and new medical scenarios.
Solution Approach 2:
The patent introduces a feedback mechanism where the auditing system continuously learns from actual prescription auditing outcomes. The machine learning model processes feedback from new prescriptions and auditing results, automatically refining and updating rules to improve adaptability while preserving the structured foundation provided by the knowledge graph.
3Measurement precision
If a prescription auditing system is introduced to assist auditing, then auditing accuracy improves, but the extent of automation and manual involvement required increases
Solution Approach 1:
The patent implements a self-improving auditing system that automatically enhances its own accuracy through machine learning. The system independently processes prescription data, learns from auditing outcomes, and automatically updates its auditing logic, thereby improving accuracy while minimizing the need for manual configuration and maintenance.
Solution Approach 2:
The patent replaces manual mechanical auditing processes with an intelligent system based on machine learning. The system automatically performs feature extraction, pattern recognition, and rule updating, substituting manual pharmacist expertise with an automated intelligent system that continuously improves its auditing accuracy without requiring proportional increases in manual involvement.
Data Source
AI summary
An information determination method and apparatus. A specific implementation solution is: acquiring prescription review information (101), wherein the prescription review information comprises: prescription information issued by a doctor and review information 5 given by a pharmacist according to the prescription information; performing feature extraction on the prescription information and the review information to generate a feature data set corresponding to the prescription review information (102); and determining the feature data set according to current review rules to obtain review results corresponding to the prescription review information (103), wherein the review rules are used to characterize a correspondence between the feature data set and the review results, and the review rules are updated on the basis of training results of a classification decision model obtained by training. The solution implements a method for determining the prescription review information by using the review rules obtained by learning, and improves the review efficiency and accuracy of a prescription review system.