Automated Vertebral Fracture Detection in CT Scans
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Vertebral compression fractures are often underdiagnosed in CT scans due to the increased workload on radiologists and the difficulty in detecting fractures in axial slices, which can lead to inadequate clinical attention, given the importance of these fractures in osteoporosis-related morbidity and mortality.
Innovation Solution
A computer-implemented method processes medical image data to determine the positions and deformity of vertebral bones, using anatomical feature labeling, segmentation, and Mahalanobis distance calculations to quantify vertebral compression fractures, and provides a mineral bone density value and fracture classification based on the Genant method standards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radiologists manually assess vertebral fractures in CT scans, then diagnostic accuracy can be maintained, but workload increases and detection efficiency decreases
Solution Approach 1:
The patent introduces an automated image processing system as an intermediary between the CT scan data and the radiologist's assessment. The system performs preliminary analysis including vertebral segmentation, deformation detection, and fracture classification, providing processed results to the radiologist for verification and final diagnosis, thus reducing manual workload while maintaining diagnostic accuracy
Solution Approach 2:
The patent replaces the manual mechanical assessment process with an automated computational system. The system uses algorithms to detect vertebral deformities, calculate deformation scores, and classify fractures automatically, substituting the radiologist's manual visual inspection with machine-based analysis to improve efficiency
2Reliability
If radiologists focus on other medical conditions such as lung cancer and COPD, then those conditions are better diagnosed, but vertebral fracture detection decreases
Solution Approach 1:
The patent segments the CT image analysis into multiple independent processing streams: one for vertebral assessment and another for other medical conditions. The vertebral analysis stream automatically performs segmentation, deformation detection, and fracture classification, allowing radiologists to focus on other conditions while vertebral fractures are assessed autonomously
Solution Approach 2:
The patent enables the vertebral fracture assessment system to perform self-service by automatically detecting, analyzing, and reporting vertebral abnormalities without requiring radiologist intervention. The system independently completes the full assessment workflow including image processing, deformation scoring, and fracture grading
3Area of stationary object
If CT images are presented in axial slices, then comprehensive anatomical coverage is achieved, but vertebral fracture detection becomes difficult
Solution Approach 1:
The patent transforms the axial slice representation into a sagittal view by reorienting and reformatting the CT image data. This dimensional transformation allows vertebral bodies to be visualized in their natural longitudinal alignment, making fractures and deformities much easier to detect while preserving complete anatomical coverage through the reconstruction process
Data Source
Figure 1~2
Figure 3
AI summary
The invention relates to a method of processing a medical image, the method comprising: receiving medical image data, the medical image data representing a medical image of at least a portion of a vertebral column; processing the medical image data to determine a plurality of positions within the image, each of the plurality of positions corresponding to a position relating to a vertebral bone within the vertebral column; and processing data representing the plurality of positions, to determine a degree of deformity of at least one vertebral bone within the vertebral column.