Medical Information Processing Apparatus for Interpretation Report Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the medical field, existing techniques for creating interpretation reports from medical images are burdensome and time-consuming, especially when dealing with numerous anatomical sections and lesions, leading to delays in image reading and report creation after image capture.
Innovation Solution
A medical information processing apparatus that associates pre-defined position information in a model image with medical information, allowing for rapid identification and display of relevant information as selection candidates based on a specified focus position, enabling efficient creation of interpretation reports without extensive image analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If image analysis is performed to detect abnormalities and create interpretation reports, then the report creation can be automated, but the processing time increases significantly when there are many anatomical sections and lesions
Solution Approach 1:
The patent applies preliminary action by pre-defining anatomical sections and lesions in a model image before actual image analysis. The system stores pre-established correspondence between model image regions and medical information in advance, so when a medical image is processed, the system only needs to match positions rather than performing comprehensive analysis, thereby reducing processing time while maintaining automation.
Solution Approach 2:
The patent segments the complex task of interpretation report creation into two parts: (1) pre-segmentation of anatomical sections and lesions in a model image with stored correspondence information, and (2) position matching between the model image and actual medical image. This segmentation allows the system to avoid re-performing comprehensive analysis for each new image, thus reducing processing time while maintaining automated report creation.
2Measurement precision
If comprehensive image analysis is performed to cover all anatomical sections and lesions, then the accuracy of interpretation reports improves, but the workload and processing time increase
Solution Approach 1:
The system performs preliminary action by pre-identifying and storing correspondence between anatomical sections/lesions in the model image and their position information before actual processing. When a medical image is received, the system only needs to match positions using the pre-established correspondence, maintaining accuracy without requiring comprehensive re-analysis, thus improving productivity.
Solution Approach 2:
The patent uses a model image as a copy or template that contains pre-defined anatomical sections and lesions with their corresponding position information. Instead of performing comprehensive analysis on each new medical image, the system uses this pre-prepared model as a reference template, copying the analytical framework to quickly identify and match features in actual images, thereby maintaining accuracy while improving efficiency.
3Measurement precision
If image analysis is performed to narrow down anatomical sections and abnormal shadows, then the interpretation report can be more precise, but the processing time increases causing delays
Solution Approach 1:
The system performs preliminary action by pre-identifying anatomical sections and abnormal shadows in a model image and storing their position information correspondence before actual image processing. When a medical image is processed, the system only needs to match positions using the pre-established correspondence, achieving precise anatomical identification without the time-consuming comprehensive analysis, thus reducing processing time while maintaining precision.
4Adaptability or versatility
If physicians manually create interpretation reports by inputting multiple items, then the report can be comprehensive, but the workload on physicians increases greatly
Solution Approach 1:
The patent applies self-service by having the system automatically generate interpretation reports based on position matching between the model image and medical image. The system autonomously identifies anatomical sections and lesions, retrieves corresponding medical information, and creates reports without requiring physicians to manually input multiple items, thereby reducing workload while maintaining report comprehensiveness through the pre-established correspondence database.
Solution Approach 2:
The system uses a model image as a template that contains pre-defined anatomical sections and lesions with their corresponding medical information. Instead of requiring physicians to manually create comprehensive reports, the system copies this pre-prepared framework and automatically matches it with actual medical images, generating comprehensive reports with minimal human intervention and significantly reducing physician workload.
Data Source
AI summary
Disclosed is a medical information processing apparatus including a medical information storage unit in which a piece of position information in a model image of a human body and a piece of medical information are stored, a first display control unit which displays a medical image, a specifying unit which specifies a focus position in the medical image, an extraction unit which converts a piece of position information of the focus position into a piece of position information in the model image and which extracts a piece of medical information corresponding to the piece of position information in the model image from the medical information storage unit, a second display control unit which displays the piece of medical information as a selection candidate, a selection unit which selects a piece of medical information, and a creation unit which creates an interpretation report by using the piece of medical information.