Medical Image Classifier Using Folding Technique for Stroke Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detection of subtly discriminated regions in volumetric medical imaging, such as early ischemic signs in non-contrast CT for acute stroke, is challenging due to the proximity of bone and subtle intensity and texture changes, leading to difficulties in distinguishing between normal and abnormal features like dense vessels and ischemia.
Innovation Solution
A medical image data processing apparatus and method that trains a classifier by selecting and aligning different parts of medical imaging data representative of the same subject, using a folding technique to create mirrored intensity channels and incorporating anatomical context for improved detection of abnormalities like thrombus or ischemia.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a naïve classifier is used to detect subtle stroke signs, then the classification process is simple, but normal calcification of arteries is confused with abnormal dense vessel signs, leading to false positives
Solution Approach 1:
The patent applies dimensionality change by folding the 3D volumetric data along the midline to create mirrored intensity channels, transforming the detection problem from a single-view classification into a multi-dimensional comparison task that distinguishes symmetric normal structures from asymmetric pathologies
Solution Approach 2:
The patent exploits asymmetry by training the classifier to recognize that normal anatomical structures are symmetric across the midline, while pathological changes create asymmetric patterns. The folding technique emphasizes this asymmetry by creating mirrored views for comparison
2Productivity
If traditional classification methods are used, then the processing is fast, but subtle intensity and texture changes in ischemia and infarcts are not detected
Solution Approach 1:
The patent applies preliminary action by performing data folding and creating mirrored intensity channels before the classification step. This preprocessing transforms the raw volumetric data into a form that emphasizes subtle differences, enabling the classifier to detect early ischemic changes more effectively without sacrificing processing speed
3Measurement precision
If detailed analysis of subtle regions is performed, then detection accuracy improves, but the time required for analysis increases
Solution Approach 1:
The patent applies self-service by using the data's own symmetry properties to facilitate detection. The folding technique automatically creates mirrored views that highlight asymmetric pathologies, allowing the classification system to efficiently identify subtle abnormalities without requiring extensive manual analysis or additional processing time
Data Source
AI summary
A medical image data processing apparatus comprises processing circuitry configured to: receive a plurality of sets of medical imaging data; and train a classifier for use in classification, wherein the training of the classifier comprises, for each of the plurality of sets of medical imaging data: selecting a first part and a second part of the respective set of medical imaging data, wherein the first part and the second part are representative of different regions of the same subject; and training the classifier for use in classification based on the first part of the set of medical imaging data and the second part of the set of medical imaging data.