AI Medical AR Image Alignment Without Surgical Markers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality applications in surgical environments require markers to accurately match medical images with patient bodies, causing disturbances and inconveniences.
Innovation Solution
A method and system using artificial intelligence to transform pre-operative medical images based on patient posture information during surgery, enabling accurate matching of medical images to real-time body images without markers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a marker is used to match medical images with patient's body in augmented reality, then the matching accuracy is improved, but the surgical environment is disturbed and convenience is reduced
Solution Approach 1:
The patent removes the marker component from the system entirely. Instead of using markers for matching, the system extracts and uses natural anatomical features from the patient's body and corresponding features from medical images to establish matching points, thereby eliminating the disturbance caused by markers while maintaining matching accuracy
Solution Approach 2:
The patent creates a virtual copy of the patient's anatomical structure by overlaying transformed medical images onto the real patient's body. This virtual model is matched to the real body using AI-based feature recognition, replacing the need for physical markers while achieving accurate spatial correspondence
2Productivity
If medical images are captured before operation and need to be matched with real-time body images, then the workflow is simplified, but the accuracy of matching is reduced due to posture changes
Solution Approach 1:
The patent implements dynamic posture estimation that continuously adapts to the patient's current posture during surgery. The system estimates posture parameters in real-time from the patient's body image and uses these dynamic parameters to transform the pre-operative medical images, ensuring they remain accurately aligned despite posture changes
Solution Approach 2:
The patent transforms medical images by applying parameter changes based on estimated posture information. The system adjusts transformation parameters such as rotation angles, translation vectors, and scaling factors according to the patient's current posture, thereby maintaining accurate matching between pre-operative images and real-time body images
Data Source
AI summary
Provided are a method and device for providing a medical augmented reality image using artificial intelligence, and more particularly, provided are a method and device for providing a medical augmented reality image by changing a medical image of a patient captured before an operation according to posture information of the patient during the operation using an artificial intelligence model and matching the changed medical image to a body image of the patient to be output to medical staff during the operation.