To develop a laboratory-based tooth model of simulated blood flow in teeth and evaluate it using ultrasound Doppler flowmetry (UDF).A laboratory-based tooth model for UDF was created based on a microfluidic experimental model proposed by Kim & Park (2016 a,b). Twenty-one maxillary or mandibular anterior human teeth within 1 month of extraction were used. Four holes were made in each tooth to fit 1.6-mm diameter polytetrafluoroethylene (PTFE) tubes: at the apical foramen, palatal surface in the centre of the crown, palatal surface apical to the cementoenamel junction (CEJ) and the root centre. Fluid mimicking pulsating blood was pumped (pressure range: 0-200 mbar, flow rate range: 0-80 μL min ) into the apical foramen via the PTFE tubes, which exited the tooth through the palatal surface in the centre of the crown (control group), palatal surface below the CEJ (group 1) and the palatal surface at the mid-root level (group 2). An UDF transducer of 20 MHz was placed at a 60° angle to the labial surface of tooth and was used to measure the fluid flow velocity (Vs, Vas, Vm, Vam, Vd, Vad and Vakd). The flow velocity of the different groups was compared using the Wilcoxon signed-rank test, with a 95% confidence level.UDF facilitated the detection of the simulated pulpal blood flow in the control group and group 1, but not in group 2. The mean and standard deviations of Vas, Vam and Vakd were 0.921 ± 0.394, 0.479 ± 0.208 and 0.396 ± 0.220 cm s , respectively, in the control group, and 0.865 ± 0.368, 0.424 ± 0.215 and 0.487 ± 0.279 cm s , respectively, in group 1. The pulpal blood flow values of the control group and group 1 were not significantly different (p > 0.05).This laboratory study revealed that ultrasound Doppler flowmetry enabled the detection of simulated blood flow below the level of the CEJ but not at the mid-root level.