
Research on Protein Level in Medical Latex Glove Images using Color Kernel Regression Method
Author(s) -
Chean Khim Toa,
K. S. Sim
Publication year - 2019
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.f1123.0886s19
Subject(s) - sample (material) , kernel (algebra) , artificial intelligence , computer science , computer vision , significant difference , graph , statistics , mathematics , chemistry , chromatography , combinatorics , theoretical computer science
In the healthcare environment, medical latex gloves are a necessary medical item for healthcare workers as it offers excellent hand barrier protection against dangerous microorganism. However, if the healthcare workers repeated exposure to the latex gloves which contain high protein level, it will increase the possibility of the workers to have a risk for latex allergy. Thus, the objective of this project is to develop a color kernel regression (CKR) method for estimating protein level through the analyses of color difference in glove images. Initially, the gloves will go through an uncomplicated chemical test for protein detection. A blue color will appear on the surface of a glove sample that contains protein. After that, the chemical binded sample will be digitally converted into a sample image using the flatbed scanner. The image will then undergo image processing to improve its quality and to calculate the color difference values of the sample. Those calculated values with the pre-defined protein levels will be used to plot a standard graph. A high coefficient of determination with R2 > 98% has been obtained from the experimental graph. This indicates that the proposed CKR method contributes significantly toward the estimation of protein level