Multi-point Annotation Using Haptic Plane in 3D Medical Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Interpreting medical images is challenging due to difficulties in distinguishing abnormal features from normal anatomical features, especially when images are acquired from different perspectives or orientations, leading to inaccurate evaluations and potential false positives or negatives.
Innovation Solution
A computer-based technique that translates and rotates a two-dimensional plane within a three-dimensional medical image to align it parallel to a reference plane, allowing for accurate annotation of anatomical structures through haptic interaction, thereby improving the interpretation of medical images by facilitating detailed multi-point annotation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If medical images are evaluated from different perspectives or orientations, then more anatomical features can be observed, but the accuracy of evaluation decreases due to difficulty in distinguishing abnormal features from normal features
Solution Approach 1:
Instead of rotating the 3D image volume itself, the patent inverts the approach by rotating the 2D display plane relative to the fixed 3D image data. The display plane can be rotated to any angle to present different anatomical perspectives, while the underlying 3D image remains stationary, making it easier to align and compare features across different orientations.
Solution Approach 2:
The patent introduces a fourth dimension by adding rotational freedom to the 2D display plane. Rather than being constrained to fixed axial, coronal, or sagittal views, the display plane can rotate freely, allowing physicians to view anatomical structures from any angular perspective while maintaining precise spatial relationships.
2Measurement precision
If the 3D image is continuously rotated and translated to accommodate different annotation planes, then accurate annotation can be achieved, but the complexity of the system increases
Solution Approach 1:
The patent extracts the rotational and translational transformation operations from the annotation process itself. By pre-aligning the 2D display plane with the desired annotation plane through automated transformation, the system removes the complexity of manual 3D manipulation while preserving annotation accuracy. The complex transformations are performed automatically based on plane parameters.
Solution Approach 2:
The system performs preliminary alignment transformations before the annotation process begins. The 3D image is pre-rotated and translated to match the orientation of the target annotation plane, so that when annotation starts, the image is already in the correct position and orientation, eliminating the need for continuous adjustments during annotation.
3Loss of information
If multiple marker points are annotated on anatomical structures, then detailed measurement information can be obtained, but the time required for annotation increases
Solution Approach 1:
The patent replaces manual mechanical manipulation of 3D images with automated computational transformations. The system automatically calculates and applies the necessary rotations and translations to align multiple annotation planes, eliminating the time-consuming manual adjustment process while enabling detailed multi-point annotation across different anatomical locations.
Data Source
AI summary
During an analysis technique, a three-dimensional (3D) image of a portion of an individual is iteratively transformed to facilitate accurate determination of detailed multi-point annotation of an anatomical structure. In particular, for a given marker point, in response to receiving information specifying a two-dimensional (2D) plane having an angular position in the 3D image, the 3D image is translated and rotated from an initial position and orientation so that the 2D plane is presented in an orientation parallel to a reference 2D plane of a display. Then, after annotation information specifying the detailed annotation in the 2D plane of the given marker point is received, the 3D image is translated and rotated back to the initial position and orientation. These operations may be repeated for one or more other marker points.