(Tuesday, 7/16/2019) 1:15 PM - 1:45 PM
Room: Exhibit Hall | Forum 9
Purpose: To enhance the greyscale-based image contrast and maximize the visibility of targeting objects by automatically adjusting display windows settings in accordance with human visual properties. And to verify the feasibility to achieve the consistent image impression to the observer by mitigating the effects of different scan conditions/personal experience/subjective preferences on image impression with optimized window settings.
Methods: An interactive way is introduced to capture the observer`s targeting object and background. By applying region-growing algorithm based on seed pixels selected by observer, the image ROI with reading focus can be determined. In accordance with human visual properties reflecting the sensitivity of human eye with different grayscale, the pixel value of object and background is mapped to the greyscale intensity range which human eyes are most sensitive to. Then the optimized display window setting can be determined based on the relation between image pixel value and display grayscale,
Results: By applying propose method, the display grayscale intensities which reflect targeting object and background are enhanced to be most effectively distinguished by human eye. The visibility of objects is maximized and the grayscale-based image contrast is enhanced. Meanwhile, the enhancement is always obtained in conformity with human visual characteristics so that the effects of different scan conditions on visual impression are mitigated. Comparing manually adjusting, automatic adjustments is time-saving as well as labor-saving. Furthermore, it is effective in achieving the consistent image impression to the observer because the deviation between different users with personal experience or preferences is eliminated to a large degree.
Conclusion: With optimized window settings in accordance with human visual properties, the grayscale-based image contrast is enhanced, the visibility of objects is maximized, and the image impression is maintained as some level of consistency, which potentially would improve the reading performance or diagnostic quality as standard procedures.
Funding Support, Disclosures, and Conflict of Interest: Authors are employees from Siemens Shanghai Medical Equipment Ltd.(Siemens Healthineers)
Keywords
Contrast, Image Visualization, CT
Taxonomy
IM- CT: Display Technology & Evaluation
Contact Email
