Objective To investigate the association between cerebral hemodynamic changes and neonatal hyperbilirubinemia. Methods A total of 67 neonates with hyperbilirubinemia who were admitted to our hospital from January 2014 to November 2015 were enrolled as study group, and another 36 normal healthy neonates were enrolled as control group. The two groups were compared in terms of end-diastolic velocity ( Vd) , systolic peak velocity ( Vs) , mean blood flow velocity ( Vm) , resistance index ( RI) , and pulsatility index ( PI) , and the serum level of bilirubin and systemic symptoms were observed after treatment. The t-test was used for comparison of continuous data between groups, and a Pearson correlation analysis was also performed. Results At the time of enrolment and on day 3 of treatment, the control group had significantly lower Vd, Vs, and Vm than the study group ( before treatment: t = 75. 873, 81. 589, 64. 600, 19. 834, 30. 453; day 3 of treatment: t = 39. 476, 55. 729, 35. 274, 6. 069, 9. 382, all P < 0. 001) . The study group had improvements in Vd, Vs, Vm, RI, and PI on day 3 of treatment. On day 5 of treatment, there were no significant differences in hemodynamic parameters between the two groups ( all P > 0. 05) . Serum level of bilirubin was positively correlated with Vd ( r = 0. 387, P < 0. 001) , Vs ( r = 0. 483, P < 0. 001) , and Vm ( r =0. 412, P < 0. 001) and negatively correlated with RI ( r =-0. 492, P < 0. 001) and PI ( r =-0. 497, P < 0. 001) . Conclusion Serum level of bilirubin interacts with cerebral hemodynamics, and cerebral hemodynamic parameters can provide objective evidence for evaluating disease progression and prognosis of neonatal hyperbilirubinemia.