This study aimed to determine the correlation of transcutaneous bilirubin (TcB) with total serum bilirubin (TSB) in infants receiving phototherapy and the accuracy of TcB for deciding the time to discontinue phototherapy. Paired TcB and TSB levels were assessed in 52 term and 56 late-preterm infants receiving phototherapy. TcB was measured in the unexposed skin of the middle forehead using a noninvasive bilirubin analyzer, BiliChek. TSB was measured every morning, and TcB was measured 2 hours after TSB measurements. Bland-Altman plots, linear regression, multivariate pair wise correlation, and receiver operating characteristic curve (ROC) analyses were performed. Bland-Altman plots before, under, and 24 hours after phototherapy in term and late-preterm infants revealed that TcB and TSB have higher consistency. TcB and TSB showed positive linear correlation before, under, and after phototherapy in both infant groups. In the multivariate pair wise correlations, differences between TcB and TSB were negatively correlated with phototherapy duration in term (correlation coefficient = −0.233, P value = .001) and late-preterm (correlation coefficient = −0.198, P value = .002) infants. Cutoff levels of TcB for discontinuing phototherapy based on the ROC analysis were 13.8 (sensitivity of 90%, specificity of 84%, area under the curve [AUC] 0.94) and 11.8 mg/dL (sensitivity of 84%, specificity of 88%, AUC 0.92) in term and late-preterm infants, respectively. TcB and TSB before, under, and after phototherapy in both term and late-preterm infants showed good correlation and higher consistency with jaundice. To reduce repetitive blood sampling for TSB, TcB measurement may be a reliable method for term and latepreterm infants undergoing phototherapy.