Medical Imaging Positioning With AI Angle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging systems lack accurate and efficient methods for positioning subjects in specific angles, relying on visual measurements and operator experience, which are time-consuming and imprecise, especially for diagnosing certain types of trauma.
Innovation Solution
A positioning assistance method using image capture apparatuses to obtain optical and depth image data, determine direction indication lines for anatomical regions of interest, and calculate included angles between these regions and imaging system hardware or between multiple regions, employing AI models like mask R-CNN and OpenPose for precise angle determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual measurement methods are used for positioning, then the operator experience can be leveraged, but the positioning precision and time efficiency deteriorate
Solution Approach 1:
The patent replaces manual visual measurement methods with an automated image processing system. The system uses image capture apparatuses to obtain optical and depth images, then employs AI models (mask R-CNN for anatomical region detection and OpenPose for pose estimation) to automatically calculate positioning angles. This substitution of mechanical/manual measurement with automated computational methods resolves the contradiction by achieving both high precision and time efficiency.
Solution Approach 2:
The system enables self-service positioning by automatically detecting anatomical regions and calculating positioning angles without requiring operator intervention for manual measurement. The AI models process the captured images autonomously to determine the required positioning parameters, allowing the system to serve itself rather than relying on operator skill and time for each measurement.
2Measurement precision
If manual positioning methods are used, then the system complexity is low, but the measurement precision and diagnostic accuracy deteriorate
Solution Approach 1:
The patent introduces AI models as intermediary components between the image capture apparatus and the positioning determination. The mask R-CNN model detects anatomical regions of interest, and the OpenPose model estimates pose information, which are then used to calculate positioning angles. These intermediary AI models enable high-precision angle measurement while managing system complexity through modular integration of specialized detection and estimation functions.
3Reliability
If quantitative angle calculation is implemented, then the diagnostic accuracy improves, but the device complexity increases
Solution Approach 1:
The patent segments the positioning assistance system into distinct functional modules: image capture apparatuses for data acquisition, mask R-CNN model for anatomical region detection, OpenPose model for pose estimation, and positioning angle calculation units. This segmentation allows each component to specialize in a specific task, improving overall reliability through coordinated quantitative calculation while managing complexity through modular architecture where each module can be independently optimized and validated.
Data Source
AI summary
Provided in embodiments of the present application are a positioning assistance method and a medical imaging system. The positioning assistance method includes: obtaining image data captured via an image capture apparatus and including a subject, the image data including optical image data and depth image data; determining, according to the image data, a direction indication line representing at least one anatomical region of interest; and calculating a body region inclination angle according to the direction indication line, the body region inclination angle including at least one of a first included angle and a second included angle, the first included angle representing an included angle between an anatomical region of interest and a plane on which hardware of a medical imaging system is located, and the second included angle representing an included angle between two anatomical regions of interest.