Automatic Pelvis Unfolding from 3D CT Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for visualizing pelvic bones in high-resolution CT scans are tedious and error-prone due to the complex structure of the pelvis, requiring radiologists to scroll through numerous axial slices, which is time-consuming and inefficient.
Innovation Solution
The method automatically detects and segments the pelvis in 3D CT images, unfolding the pelvis anatomy to generate a reformatted image that provides improved visualization, allowing for quicker identification of abnormalities like metastases and fractures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radiologists scroll through all axial slices in a CT volume to examine the pelvis, then complete evaluation of pelvic bone can be achieved, but the examination process becomes tedious and time-consuming (averaging 14 minutes)
Solution Approach 1:
The patent segments the pelvic bone from the entire CT volume by extracting only the relevant pelvic region, separating it from the surrounding abdominal structures. This segmentation allows radiologists to focus exclusively on the pelvis without scrolling through all axial slices, thereby reducing examination time while maintaining evaluation completeness.
Solution Approach 2:
The patent transforms the 3D CT volume into a 2D unfolded representation of the pelvic bone surface. By flattening the complex 3D pelvic structure into a 2D map, the system enables rapid visual assessment of the entire pelvis in a single view, eliminating the need to scroll through multiple axial slices while preserving diagnostic accuracy.
2Reliability
If radiologists manually scroll through axial slices to locate abnormalities, then diagnostic accuracy can be maintained, but the process becomes error-prone and inefficient
Solution Approach 1:
The patent performs preliminary automated segmentation and unfolding of the pelvic bone before the radiologist begins examination. The system pre-processes the CT data to extract and flatten the pelvic surface, so that when the radiologist views the unfolded pelvis, the entire structure is already optimized for visual assessment, eliminating the need for manual scrolling and improving both efficiency and accuracy.
Solution Approach 2:
The patent creates a simplified 2D copy or representation of the 3D pelvic bone structure through unfolding. This 2D unfolded view serves as an optimized copy that preserves all diagnostic information while presenting it in a format that is easier and faster to interpret, thereby improving examination efficiency without compromising diagnostic accuracy.
3Measurement precision
If the pelvis is visualized in standard axial CT slices, then anatomical detail is preserved, but the complex pelvic structure makes visualization difficult and time-consuming
Solution Approach 1:
The patent transforms the complex 3D pelvic anatomy into a 2D unfolded representation. This dimensional transformation simplifies the visualization by flattening the complex curved surfaces of the pelvis into a planar map, making it easier for radiologists to interpret the anatomy while preserving all necessary structural details through accurate geometric transformation.
Solution Approach 2:
The patent changes the geometric parameters of the pelvic visualization by applying coordinate transformation and unfolding algorithms. These parameter changes reconfigure the 3D pelvic surface into a 2D format with adjusted geometry, maintaining anatomical fidelity while improving visual interpretability and reducing the cognitive load on radiologists during examination.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method and system for automatic pelvis unfolding from 3D computed tomography (CT) images is disclosed. A 3D medical image, such as a 3D CT image, is received. Pelvis anatomy is segmented in the 3D medical image. The 3D medical image is reformatted to visualize an unfolded pelvis based on the segmented pelvis anatomy.