Medical Image Motion Correction via Target Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for correcting respiratory motion in medical images are inaccurate, affecting the analysis of target organs by not properly isolating their motion from coupled organ motions.
Innovation Solution
A computer-implemented method for medical image processing that involves acquiring multiple frames of medical data, performing image segmentation to obtain a mask for a target region, and using this mask for motion correction of floating frames based on correlation information and iterative processing to achieve accurate motion correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional motion correction methods (rotation, translation, miscut, time or space angle) are used to correct respiratory motion of other organs, then the respiratory motion of other organs can be corrected, but the motion correction image is inaccurate
Solution Approach 1:
The patent segments the image into target region and background region using image segmentation and mask generation. This allows separate processing of the target organ (which should maintain its respiratory motion) from other organs (which need motion correction), thereby improving motion correction accuracy without compromising analysis reliability
Solution Approach 2:
The patent applies different processing strategies to different regions: the target region maintains its original respiratory motion characteristics while the background region undergoes motion correction. This local differentiation resolves the contradiction by ensuring accurate motion correction for background organs while preserving the reliability of target organ analysis
2Measurement precision
If motion correction is applied to correct respiratory motion of other organs, then other organ motions are corrected, but the target organ motion is affected
Solution Approach 1:
By segmenting the image into target region and background region, the patent enables selective motion correction that affects only the background region while leaving the target region unchanged, thus correcting other organ motions without distorting target organ motion
Solution Approach 2:
The patent extracts the target region from the full image using mask generation, isolating it from the motion correction process. This extraction allows the motion correction to be applied only to the background region, preventing any harmful effects on the target organ motion
Data Source
AI summary
A method for medical image processing is disclosed. The method includes: acquiring a plurality of frames of medical image data including a reference frame and at least one floating frame, each of the reference frame and the floating frame including a target region and a background region; performing image segmentation on the reference frame to obtain a mask for the target region in the reference frame; performing, based on the reference frame and the mask, motion correction on the floating frame to obtain a motion corrected image corresponding to the floating frame, and performing motion analysis on the target region in the motion corrected image. A computer apparatus and a non-volatile computer readable storage medium for medical image processing are also disclosed.