Medical Image Classification Model for PACS Hanging Protocols
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Commercial picture archive and communication systems (PACS) face challenges in accurately differentiating between various classifications of medical images, such as anatomical and textual images, which interferes with the application of hanging protocols and affects the display of medical exams, leading to irrelevant image interruptions and user fatigue.
Innovation Solution
A system and method that utilize an electronic processor to classify medical images using a classification model, determining whether an image is anatomical or textual, and applying a set of rules to display images accordingly, ensuring relevant and efficient image presentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If PACS uses rule-based hanging protocols to automatically present medical images, then image presentation efficiency is improved, but image classification accuracy deteriorates due to insufficient metadata information
Solution Approach 1:
The patent introduces an image classification model as an intermediary component between the image database and the hanging protocol system. This model analyzes image content to generate classification labels (anatomical vs. textual), which then inform the hanging protocol's image selection and arrangement decisions. The intermediary resolves the contradiction by providing accurate classification information that the original metadata-limited system could not obtain.
2Quantity of substance
If PACS includes all images from an exam in the display, then completeness of information is improved, but user experience deteriorates due to irrelevant textual image interruptions
Solution Approach 1:
The patent segments the image set into distinct categories (anatomical images and textual images) based on classification results. The hanging protocol then applies different display rules to each segment: anatomical images are displayed according to medical review requirements, while textual images are filtered out or displayed separately. This segmentation allows the system to maintain information completeness for anatomical images while eliminating interruptions from textual images.
3Loss of information
If PACS displays anatomical and textual images together, then all available data is presented, but interpretation speed deteriorates due to irrelevant image interruptions
Solution Approach 1:
The patent applies different quality characteristics to different image types through classification-based display rules. Anatomical images receive full display priority with optimized presentation characteristics, while textual images are assigned lower priority or alternative display characteristics. This local quality differentiation ensures that anatomical images maintain their interpretability and viewing speed without being disrupted by textual content.
Data Source
AI summary
Methods and systems for displaying a medical image captured as part of a medical imaging procedure. One system includes an electronic processor configured to receive a plurality of images from an image database, the plurality of images included in a medical exam associated with the procedure. The electronic processor is also configured to automatically determine a classification for each of the plurality of images using a classification model analyzing content of each of the plurality of images, the classification including one of a textual image and an anatomical image. The electronic processor is also configured to determine a set of rules displaying the plurality of images and display a subset of the plurality of images based on the classification determined for each of the plurality of images and the set of rules.