Learned Capture-Condition Selection for Medical Image Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image diagnosis examinations face challenges in dynamically determining an appropriate image capturing condition based on the subject's state, leading to inefficient radiation exposure and time consumption, especially in invasive procedures.
Innovation Solution
A medical image diagnosis apparatus that includes an acquisition unit and an image capturing condition determination unit, which utilizes a learned model to dynamically determine the image capturing condition based on subject information and evaluation information, optimizing image capturing for improved image quality and reduced uncertainty in disease classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a wide image capturing range is set to ensure sufficient examination coverage, then diagnostic completeness is improved, but radiation exposure and examination time increase
Solution Approach 1:
The system dynamically adjusts the image capturing range to match the actual diagnosis target region. Instead of uniformly capturing the entire examination range, the learned model identifies and prioritizes capturing only the regions containing diagnosis targets, thereby reducing radiation exposure to non-target areas while maintaining diagnostic completeness for target regions.
Solution Approach 2:
The image capturing condition determination unit dynamically determines the image capturing range based on real-time analysis of medical images and subject information. The system adapts the capturing range throughout the examination process, expanding or contracting it according to the detected presence and position of diagnosis targets, rather than using a fixed predetermined range.
2Reliability
If a wide image capturing range is set to ensure sufficient examination coverage, then diagnostic completeness is improved, but examination time increases
Solution Approach 1:
The system focuses imaging resources on specific regions containing diagnosis targets rather than uniformly imaging the entire examination range. This localized approach reduces the total volume of data to be processed and displayed, thereby shortening examination time while maintaining diagnostic completeness for target regions.
Solution Approach 2:
The system dynamically adjusts the image capturing range during the examination based on detected diagnosis targets. When targets are identified, the system concentrates on capturing and processing only relevant regions, reducing overall examination time compared to static full-range imaging protocols.
3Manufacturing precision
If manual determination of image capturing condition is used to optimize for each subject, then image capturing appropriateness is improved, but operational complexity increases
Solution Approach 1:
The system performs automatic determination of image capturing conditions using the learned model that analyzes subject information and medical images. The determination of diagnosis target presence, position, and appropriate capturing range is executed autonomously without requiring manual user input, thereby maintaining high appropriateness while reducing operational complexity.
Solution Approach 2:
The manual judgment and decision-making process is replaced by an automated learned model that processes medical images and subject information. This substitution of human cognitive operations with an automated system maintains the precision of expert-level determination while eliminating the operational burden and complexity for users.
4Object-affected harmful factors
If dynamic determination of image capturing condition is implemented, then radiation exposure is reduced, but system complexity increases
Solution Approach 1:
The system performs preliminary analysis using the learned model to predict the presence and position of diagnosis targets before finalizing the image capturing range. This advance determination allows the system to configure optimal capturing conditions that minimize radiation exposure from the outset, rather than requiring complex real-time adjustments during the examination.
Solution Approach 2:
The learned model serves as an intermediary between the raw medical images/subject information and the image capturing condition determination. This intermediate processing layer translates complex image data into actionable insights about diagnosis targets, simplifying the overall system architecture while enabling dynamic optimization of radiation exposure.
Data Source
AI summary
According to one embodiment, a medical image diagnosis apparatus includes an acquisition unit and an image capturing condition determination unit. The acquisition unit acquires subject information regarding a subject that includes a first medical image obtained by capturing an image of a diagnosis target region of the subject under a first image capturing condition. The image capturing condition determination unit determines a second image capturing condition to capture a second medical image of the subject, based on the subject information, and a learned model obtained by executing learning while associating evaluation information regarding a medical image and an image capturing condition.