Aim: Eligibility for liver transplantation for hepatocellular carcinoma (HCC) is currently based on single-dimension, diameter measurements on cross-sectional imaging, as specified by various selection criteria. This does not account for significant differences in shape, and therefore tumour volume, between patients. This study investigated whether one-dimensional selection criteria disadvantages patients by not considering volume. Methods: Patient data were collected retrospectively from a prospectively maintained database. Tumours were measured on both computer tomography (CT) and magnetic resonance imaging (MRI). Tumour volume was measured using two methods; semi-automated planimetry and the ellipsoid volume formula. Statistical analysis was performed using SPSS. Results: A total of 313 patients with HCC were assessed for liver transplantation. For this study, patients who underwent transplantation (n = 89) and those who did not based on tumour size (n = 33) were included. In total, 213 tumours were measured, showing excellent correlation between CT and MRI (R2 = 0.83). The majority of tumour nodules (94％) were ellipsoid not spherical. Volumetric measurements of the 84 tumours that did not meet diameter-based Milan criteria confirmed that 76％ would have been within a theoretical volume allowance based on Milan criteria diameters. Conclusion: This study shows that a significant number of patients deemed outside conventional diameter-based Milan criteria have smaller tumour volumes than those considered within criteria. It appears that those with ellipsoid rather than spherical tumours may be disadvantaged by current size-based criteria. Further research using a contemporary patient cohort who have had the benefit of advancements in non-surgical treatments for HCC is required.