Medical Image Registration Using Stable Reference Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge lies in registering medical images of different modalities, particularly in real-time ultrasound images and non-real-time MRI or CT images, where the movement of a patient during a medical procedure makes it difficult to accurately align and fuse these images due to inadequate image quality and the inability of MRI or CT equipment to capture real-time data.
Innovation Solution
An apparatus and method that utilize a processor to register medical images by acquiring cross-sections of ultrasound and other modality images using a reference point, such as a bone or the belly button, which remains undeformed during respiration, and calculates transformation relationships to align the images, allowing for real-time registration and fusion of images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature points extracted from each medical image are used to register images of different modalities, then image registration can be performed, but image registration becomes difficult due to inadequate quality of medical images
Solution Approach 1:
The patent introduces a reference object (such as a patient's body part or external marker) as an intermediary between the ultrasound image and the MRI/CT image. This reference object provides stable, identifiable features that serve as a common reference frame, enabling accurate registration despite the inadequate quality and different modalities of the medical images themselves
Solution Approach 2:
The patent creates a virtual model or copy of the reference object from the high-quality MRI or CT image, then uses this virtual copy to guide the registration process with the real-time ultrasound image. This allows the system to leverage the accuracy of MRI/CT anatomy while maintaining the real-time capability of ultrasound
2Measurement precision
If MRI or CT equipment is used to obtain high-resolution images of organs and lesions, then clear distinction of organs and lesions is achieved, but real-time detection of patient's respiration or movement is not possible
Solution Approach 1:
The patent performs MRI or CT scanning before the surgical procedure to obtain high-resolution anatomical images. These pre-acquired images serve as a static reference framework that is then registered with real-time ultrasound images during the procedure, allowing the system to benefit from both high resolution and real-time capability
Solution Approach 2:
The patent dynamically adjusts the registration between the static MRI/CT images and real-time ultrasound images by tracking the position of the reference object or probe during the procedure. This allows the system to maintain accurate alignment despite patient movement or respiration, effectively creating a dynamic registration system that adapts to changing conditions
3Speed
If real-time ultrasound images are used to detect patient's respiration or movement during operation, then real-time detection is achieved, but it becomes difficult to identify the shapes of organs and lesions with a naked eye
Solution Approach 1:
The patent merges the real-time ultrasound image with the pre-acquired MRI or CT image through image registration. The ultrasound image provides real-time information about patient movement and respiration, while the MRI/CT image provides detailed anatomical structures. By combining these images in a registered coordinate system, the system simultaneously achieves real-time detection capability and accurate organ shape identification
Data Source
AI summary
An apparatus for registering medical images includes a processor configured to acquire, via an ultrasound probe, a first cross-section of a first medical image, the first cross-section comprising a reference point of an object; and a memory configured to store a second medical image. The processor is configured to obtain a second cross-section of the second medical image corresponding to the first cross-section by using the reference point, and register the first and second medical images with each other.