Method: the sample comprised 33 growing patients (mean age 10.7 years, range 7.2 - 14.5 years) and an oropharynx acrylic phantom (OAP) scanned with cone-beam computed tomography (CBCT) i-CAT Scanner (0.3 voxel resolution). The OAP physical volume was known and used as a gold standard. Automatic segmentations and volume analysis of OP and OAP DICOM data was performed using Mimics, ITK-Snap, OsiriX, Dolphin3D, InVivo Dental and Ondemand3D software. The intraclass correlation coefficient (ICC) was used for the repeatability study. The Wilcoxon test for non-parametric data was used to compare the OAP segmentation volume to the OAP actual physical volume. The ANOVA was used to compare the patient’s OP volume measurements
Result: The ICC was high for both OAP (0.99) and OP (>0.94) volume measurements. The OAP segmentation was statistically different than the OAP actual physical volume (p=0.005). The error difference in volume calculation was 0.2% for Mimics, 1% for Dolphin3D, 1.3% for OsiriX, 1.8% for ITK-Snap, 6.4% for Ondemand3D and 11% for InVivo Dental. There was a statistical difference between the 6 DICOM viewers (p<.001) for segmentation and volume analysis of OP. Higher mean volumes (mm3) were observed for Mimics (7163), ITK-Snap (7174), OsiriX (7086), Dolphin3D (7071) and lower mean volumes was observed for InVivo Dental (6661) and Ondemand3D (6061). There was no statistical difference (p<0.05) among Mimics, ITK-Snap and Dolphin3D. The OP segmentation with InVivo Dental was similar to OnDemand3D, but both were statistically different (p<.05) to the other 4 software
Conclusion: Mimics, ITK-Snap, Dolphin3D and OsiriX showed higher accuracy than InVivo Dental and Ondemand3D software for segmentation and volume analysis of the OP