Medical Image Display Consistency via Deep Neural Network Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical imaging technologies face challenges in achieving consistent display settings for medical images, leading to inconsistent presentation and complicating the evaluation process for clinicians, due to sensitivity to noise and artifacts in histogram analysis, which fails to incorporate contextual information.
Innovation Solution
The use of deep neural networks to map medical images to appearance classification cells, incorporating contextual information and user preferences, to select optimal window-width and window-center settings, thereby reducing sensitivity to noise and artifacts and achieving consistent display appearance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If histogram analysis is used to automatically set display settings, then the process is automated, but the display appearance becomes inconsistent due to sensitivity to noise, artifacts, and lack of contextual information
Solution Approach 1:
The patent introduces an appearance classification matrix as an intermediary framework between the automatic histogram analysis and the final display settings. This matrix categorizes images into discrete appearance classes based on multiple features (mean intensity, standard deviation, skewness, kurtosis, modality), serving as a mediator that translates complex image characteristics into standardized display parameter selections, thereby improving consistency while maintaining automation.
Solution Approach 2:
The patent transforms the continuous histogram data into discrete appearance classification categories by evaluating multiple statistical parameters (mean, standard deviation, skewness, kurtosis) and modality characteristics. This parameter transformation converts the sensitive continuous data into robust discrete classes, reducing the impact of noise and artifacts on display settings selection.
2Device complexity
If histogram analysis is used to determine display settings, then computation is simplified, but contextual information about anatomical regions is lost
Solution Approach 1:
The patent segments the image analysis process into distinct feature extraction components (mean intensity, standard deviation, skewness, kurtosis, modality assessment) and evaluates each separately. This segmentation allows the system to capture multiple aspects of image appearance and contextual information independently, then integrate them into the appearance classification matrix without requiring complex unified computation.
Solution Approach 2:
The patent applies different evaluation criteria to different aspects of image appearance. Specific statistical measures (mean, standard deviation) are used for certain features while modality-specific criteria are applied to assess anatomical content. This local quality approach ensures that contextual information about different image regions and characteristics is preserved and utilized appropriately in display settings selection.
3Manufacturing precision
If manual adjustment of display settings is required to achieve desired appearance, then display appearance can be optimized, but workflow efficiency decreases
Solution Approach 1:
The patent performs preliminary classification of images into appearance categories before final display settings are applied. By pre-evaluating image characteristics and assigning appearance classes upfront, the system prepares the groundwork for automatic settings selection, eliminating the need for manual adjustment while ensuring optimized display quality. This preliminary action maintains both efficiency and quality.
Solution Approach 2:
The patent implements a feedback mechanism where the appearance classification results directly inform the selection of display settings. The system continuously evaluates image appearance features, compares them against the classification matrix, and automatically adjusts settings based on this feedback loop, replacing manual adjustment with an automated quality-ensuring process that improves workflow efficiency.
Data Source
AI summary
A method for automatic selection of display settings for a medical image is provided. The method includes receiving a medical image, mapping the medical image to an appearance classification cell of an appearance classification matrix using a trained deep neural network, selecting a first WW and a first WC for the medical image based on the appearance classification and a target appearance classification, adjusting the first WW and the first WC based on user preferences to produce a second WW and a second WC, and displaying the medical image with the second WW and the second WC via a display device.