Current studentsExisting staffContact us
 

Computer-aid system for automated jaundice detection

Simple item page
dc.contributor.authorAbdulrazzak, A.Y.
dc.contributor.authorMohammed, S.L.
dc.contributor.authorAl Naji, A.
dc.contributor.authorChahl, J.
dc.date.issued2023
dc.description.abstractAt the beginning of their lives, newborns may have a widespread condition known as Jaundice or Hyperbilirubinemia. High levels of bilirubin in the blood are the primary cause of jaundice. Severe cases of jaundice may cause acute bilirubin encephalopathy due to the toxicity of bilirubin to the cells of the brain, which may lead to kernicterus. Kernicterus causes several symptoms, including a permanent upward look, loss of hearing, and repetitive and uncontrolled movements. Therefore, diagnosing this condition at the appropriate time helps to prevent chronic effects. In this study, jaundice or hyperbilirubinemia is diagnosed using a computer vision system based on a random forest algorithm. The system comprises a digital HD camera, a computer device with a Matlab application installed to analyze and detect the skin color changes of the infant, and an Arduino Uno microcontroller to control an LED ultraviolet light. A set of neonate images were collected to train the random forest algorithm, including 374 for normal and 137 for jaundiced infants. |The experimental results using the random forest algorithm for classification reached an accuracy of 98.4375%. The results of this study are promising and open doors for new monitoring applications in various medical diseases detection with a high degree of accuracy.
dc.identifier.citationJournal of Techniques, 2023; 5(1):8-15
dc.identifier.doi10.51173/jt.v5i1.1128
dc.identifier.issn2708-8383
dc.identifier.issn2708-8383
dc.identifier.urihttps://hdl.handle.net/11541.2/33595
dc.language.isoen
dc.publisherMiddle Technical University
dc.rightsCopyright 2023 The author(s). This work is licensed under a Creative Commons Attribution 4.0 International License. (https://creativecommons.org/licenses/by/4.0/)
dc.source.urihttps://doi.org/10.51173/jt.v5i1.1128
dc.subjectjaundice
dc.subjecthyperbilirubinemia
dc.subjectphototherapy
dc.subjectskin color analysis
dc.subjectrandom forest algorithm
dc.titleComputer-aid system for automated jaundice detection
dc.typeJournal article
pubs.publication-statusPublished
ror.fileinfo12261510020001831 13261510010001831 2-1128-ENGINEERING
ror.mmsid9916727921901831

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
2-1128-ENGINEERING.pdf
Size:
967.62 KB
Format:
Adobe Portable Document Format
Description:
Published version
Download

Collections