Method of quantitatively measuring fat content in target organ from ultrasound visual image
AssigneeElectronics and Telecommunications Research Institute
InventorsJi-wook Jeong, Soo-yeul Lee, Seung-hwan Kim
Life of this patent
Estimating the term…
Abstract
A method of quantitatively assessing fat content in a target organ from an ultrasound visual image is provided. This method includes obtaining an ultrasound visual image of the target organ, setting a target region in the obtained image, measuring a quantified representative gray level of the target region from a gray level distribution of pixels of the target region, and assessing fat content corresponding to the quantified representative gray level of the target region.
Claims10
A method of quantitatively assessing fat content in a target organ from an ultrasound visual image, the method comprising:
- obtaining an ultrasound visual image of the target organ;
- setting a target region and comparative regions in the obtained image;
- measuring a representative gray level of the target region from a gray level distribution of pixels of the target region;
- measuring representative gray levels of the comparative regions from the gray level distribution of pixels of the comparative regions;
- quantifying the representative gray level of the target region as relative values with regard to the representative gray levels of the comparative regions;
- and
- assessing fat content corresponding to the quantified representative gray level of the target region, wherein said target region measuring includes:
- obtaining a distribution pattern of pixels of the target region;
- obtaining histogram distributions of the number of pixels with regard to gray levels of each pixel of the target region;
- and
- determining the gray level of one histogram of the histogram distributions as the representative gray level, and
- wherein the representative gray level of the target region is determined as a gray level of a histogram positioning at a point of inflection in the histogram distribution.
Dependencies are inferred from the claim text.