Microscope Image Area Selection Using ML-Based Magnification Guidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to efficiently identify suitable areas for microscopic examination in smear objects, requiring high operator skill and burdening the operator with visual transitions between the microscope eyepiece and display unit, while failing to synthesize findings of microbes and living body components effectively.
Innovation Solution
An image processing apparatus that analyzes images captured by a microscope, specifies areas suitable for microscopic examination, and adjusts magnification automatically, using machine learning for semantic segmentation and object detection to highlight these areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an operator manually retrieves areas suitable for microscopic examination by changing magnification, then accurate identification of examination areas can be achieved, but high operator skill is required and the process becomes time-consuming
Solution Approach 1:
The patent replaces the manual mechanical process of changing magnification and visually searching for areas with an automated image processing system. The system captures images at multiple magnifications, automatically processes them to identify suitable examination areas, and presents results to the operator, eliminating the need for manual skill-based retrieval while maintaining accuracy.
Solution Approach 2:
The system performs preliminary analysis by capturing and processing images at multiple magnifications before the operator needs to examine specific areas. By pre-identifying suitable examination areas through automated image processing, the system prepares the information in advance, saving the operator time and effort during actual examination.
2Loss of information
If an operator uses both the microscope eyepiece and display unit for observation, then comprehensive examination information can be obtained, but the operator experiences visual burden from turning the visual line
Solution Approach 1:
The patent merges the information from multiple magnification views and the eyepiece observation with the display unit information into a single integrated display. The system presents synthesized examination areas that combine data from both sources, allowing the operator to obtain comprehensive information while viewing everything on one display, thereby eliminating the need to shift visual focus between the eyepiece and screen.
3Device complexity
If only classification information is displayed without synthesized evaluation, then simple presentation of sputum propriety can be achieved, but it is difficult to specify areas suitable for microscopic examination by synthesizing findings
Solution Approach 1:
The patent applies different evaluation criteria and weighting to different regions and features within the image. Instead of uniform processing, the system identifies specific areas with different characteristics (such as regions with microbes versus regions with living body components) and evaluates each according to its local requirements, thereby improving the accuracy of suitable area specification while maintaining manageable complexity through targeted analysis.
Data Source
AI summary
An image processing apparatus supports observation of an object using a microscope. The image processing apparatus includes an obtainment unit and specifying unit. The obtainment unit obtains an image captured from an object observed by using a microscope. The specifying unit specifies at least one area in the image that will be captured at an expanded magnification based on evaluation of a detection target included in the image and a magnification of the microscope in capturing the image.