Prediction Imaging for Standing Spinal Alignment Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current surgical systems struggle to predict the effect of angular changes in patient bones after surgery when the patient transitions from a predefined body posture, such as lying down, to a natural posture, like standing, due to reliance on surgeon experience and lack of accurate prediction tools.
Innovation Solution
A method and system that utilize image data and alignment data to modify medical images, applying rotational transformations and spinal balance parameters to predict bone poses in a different body posture, enabling accurate visualization of spinal balance parameters in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a surgeon relies on experience to predict surgical outcomes, then the prediction process is simple and quick, but the accuracy and reliability of the prediction deteriorates
Solution Approach 1:
The system creates a digital copy of the patient's anatomy through 3D modeling from medical images, allowing virtual simulation of surgical outcomes without physical experimentation. This digital twin approach enables accurate prediction while avoiding the need for complex physical measurement devices during surgery.
Solution Approach 2:
The system performs surgical simulation and outcome prediction before the actual surgery takes place. By pre-calculating the effects of angular changes and posture transitions, the system provides accurate predictions that guide surgical planning without adding complexity during the surgical procedure itself.
2Measurement precision
If fluoroscopic shots are used to verify surgical alignment, then measurement accuracy is improved, but radiation exposure and surgical time increase
Solution Approach 1:
The system replaces physical fluoroscopic verification with a digital copy of the surgical outcome. By comparing the pre-calculated prediction images with actual surgical results, the system achieves alignment verification without exposing the patient to additional radiation from intraoperative fluoroscopy.
Solution Approach 2:
The system substitutes the mechanical fluoroscopy imaging system with a computational imaging approach. Instead of using X-ray equipment to verify alignment during surgery, the system uses pre-acquired medical images processed through 3D modeling and image modification algorithms to predict and verify alignment.
3Measurement precision
If multiple medical images are acquired in different postures, then prediction accuracy is improved, but the time and complexity of image processing increases
Solution Approach 1:
The system performs image acquisition and processing before surgery in a preoperative planning phase. By completing all image modifications, 3D model creations, and prediction calculations beforehand, the system eliminates time-consuming processing during surgery while maintaining high prediction accuracy.
Solution Approach 2:
The system uses a unified 3D modeling approach that can process multiple types of medical images (CT, MRI, X-ray) from different postures through a single integrated workflow. This multi-functional processing framework reduces overall complexity and time compared to separate processing pipelines for each image type.
Data Source
Figure 1
Figure 2
Figure 3a~3b
AI summary
A method for determining at least one prediction image based on at least one medical image of a patient is provided. The method comprises: obtaining image data including at least one medical image comprising representations of a plurality of bones of a patient's body in a first body posture; obtaining alignment data indicative of an alignment of at least two of the plurality of bones of the patient's body in a second body posture differing from the first body posture; determining, based on the image data and the alignment data, at least one prediction image comprising representations of the plurality of bones of the patient's body in the first body posture, wherein determining the at least one prediction image comprises modifying the at least one medical image based on the alignment data such that poses of the representations of the at least two bones after the modification comply with the alignment of the at least two bones indicated by the alignment data; and triggering display of one or more of the at least one prediction image and/or of information derived from one or more of the at least one prediction image. A corresponding system, computer program and carrier are also provided.