A Mobile Computer-Aided Diagnosis of Neonatal Hyperbilirubinemia using Digital Image Processing and Machine Learning Techniques

Abstract

Neonatal Hyperbilirubinemia, or jaundice, is a harmful disease found in newborns, a symptom of which is the yellowish discoloration of the skin. Visual examination is most frequently used for screening of Hyperbilirubinemia in neonates, however, blood specimen collection is the gold standard to identify the disease and its severity. We propose a Mobile Computer-Aided Diagnosis (mCADx) tool to identify the Neonatal Hyperbilirubinemia symptom using advanced digital image processing and data mining techniques. The mCADx was developed in a cross-platform environment. The mCADx works with smart devices run on either iOS or Android operating systems. With ethical committee approval, we collected and studied image data of 178 infant subjects with different jaundice severity levels. The severity of the disease was examined from blood test results, which were annotated by medical specialists. Data mining techniques included Decision Trees, k Nearest Neighbor, and the Conventional Neural Network was investigated in the dataset. An in-depth comparison between techniques was performed and discussed. The classification results in CNN gained the highest accuracy at 0.8099, 0.9251, 0.8086. This novel work can assist in identifying Neonatal Hyperbilirubinemia in newborns after discharging from the hospital. Reoccurring Neonatal Hyperbilirubinemia can be found with minimum awareness of parents. Limitations and future works were discussed in this work.

A Non-invasive Methods for Neonatal Jaundice Detection and Monitoring to Assess Bilirubin Level: A Review

“…Different machine learning methods, such as an www.aetic.theiaer.org ensemble of different classifiers, checking KNN, LARS-Lasso elastic net, LARS, SVR, and RF, have been employed to ensure jaundice. Using methods for sophisticated digital image processing and data mining, Dissaneevate et al [47] suggested a Mobile Computer-Aided Diagnosis (mCADx) tool to diagnose the Neonatal Hyperbilirubinemia symptom. They explored data mining approaches using the dataset, including Decision Trees, k Nearest Neighbors, and the Conventional Neural Network, on picture data of 178 infant participants with varying degrees of jaundice severity.…”

Comparative Analysis of Classification of Neonatal Bilirubin by Using Various Machine Learning Approaches

“…Almost 60% of term neonates and 80% of preterm neonates develop symptoms of jaundice [ 1 ]. It is observed that when the bilirubin level exceeds 5 mg/dL, it leads to yellow discoloration of the sclera and skin, and an increase in the level may lead to Kernicterus or cause death [ 2 ]. The most common treatment for this increased bilirubin, or hyperbilirubinemia, is either phototherapy or exchange transfusion [ 3 , 4 ].…”

“…Dissaneevate et al introduced a Mobile Computer-Aided Diagnosis (mCADx) platform, employing decision trees, k-nearest neighbor, and convolutional neural networks (CNNs), with the CNN achieving the highest accuracy [ 2 ]. Sammir et al created a portable monitoring system capturing sclera images with a custom goggle, achieving a 90% accuracy rate [ 12 ].…”

Investigation of the effect of arteriole clogging of the heart on the characteristics of intravenous blood flow

“…Their results include changes in diameter and thickness with stress distribution for both monolayer and bilayer mod-els, indicating a major di erence between monolayer and bilayer models in stress distribution. In [23], the stress distribution in the abdominal aortic aneurismal wall was investigated using nite element modeling. The results of this study indicate that in addition to the size of the aneurysm, mechanical properties and blood ow may also signi cantly a ect the stress distribution in the aneurismal wall.…”