PCA- Based Feature Extraction and k-NN algorithm for Early Jaundice Detection

Abstract

Abstract. Jaundice is a yellow discoloration of the skin and/or whites of the eyes that are often seen in newborn infants. The discoloration is caused by a yellow substance called bilirubin. Infants with high blood levels of bilirubin, called hyperbilirubinemia, develop the yellow color when bilirubin accumulates in the skin. The main symptom of jaundice is yellow coloring of the skin and conjunctiva of the eyes. Jaundice can also make babies sleepy which can lead to poor feeding. Poor feeding can make jaundice worse as the baby can become dehydrated. If a baby has conjugated jaundice, it may have white chalky stool (poo) and urine that is darker than normal. A PCA method was employed to study the behavior of the infant. The experimental results reveal that the proposed method can minimize the morbidity and mortality than the conventional method based on k-NN Algorithm.