Irreversible electroporation (IRE) is a new ablation modality. Our purpose was to describe the effectiveness and the safety of the treatment and to evaluate the magnetic resonance imaging (MRI), computed tomography (CT) and contrast-enhanced ultrasound (CEUS) diagnostic accuracy in HCC patients treated with IRE at 1-, 3-, and 6-month follow-up. In an 18-month period, we treated 24 HCC lesions in 20 patients unfit for surgery. MRI, CT and CEUS were performed before and one, 3 and 6 month after IRE. We employed the liver-specific contrast medium Primovist (gadolinium ethoxybenzyl dimeglumine) in MRI. After IRE the lesions were classified as responders or non-responders to the treatment according to the mRECIST and the complications were recorded. We evaluated the size, shape, signal intensity (T1-W, T2-W, and DWI) in MRI, dynamic contrast enhancement pattern for CEUS, CT and MRI and signal behavior during the liver-specific phase for MRI. According to mRECIST, at 1 month MRI and CEUS showed a complete response (CR) in 91.7% of cases (22/24) tumors, while there was partial response (PR) in the remaining 2/24 (8.3%) treated nodules; in CT study all ablated zone appeared as necrotic (CR 100%). The residual viable tumor in MRI and in CEUS study had similar diameter (10 mm). No new HCC were identified from MRI, CT or CEUS. At 3 months MRI and CEUS showed the same results seen after 1 month from the treatment. Twenty-two necrotic lesions, and 2 residual tumors were found (CR = 91.7% and PD = 8.3%). In MRI study the two cases of residual tumor tissue had a diameter of 11 and 12 mm each. At CEUS the diameter of residual HCC was similar to the diameter at 1 month. CT showed 23 necrotic areas and one residual viable tissue in the treated zone, with a diameter of 10 mm (CR = 95.3% and PD = 4.7%). No new foci of HCC were identified from all imaging studies. At 6 months MRI, CEUS, and CT showed 22 necrotic lesions and 2 residual tumors in ablated zone (CR = 91.7% and PD = 8.3%). At MRI the diameters of the two residual viable HCCs were 12 and 14 mm, at CEUS the diameters were 11 and 12 mm, while at CT the diameters were 10 and 10 mm. No statistical difference was evaluated between CR, PR, PD percentage values for MRI, CT and CEUS (p value > 0.05 at Chi-square test). No major vascular complication was recorded after IRE. Six out of 20 patients (30%) showed a transient hepatic intensity difference (THID) area within the normal liver parenchyma adjacent to the treated lesions. Two of the 20 patients (10%) showed an absent concentration of liver-specific contrast medium around the ablation zone. Two patients developed complications, consisting in a peripheral arteriovenous shunt and a segmental dilation of the intrahepatic biliary ducts. We found no statistically significant difference in morphology, size (variation in the largest diameter), signal intensity in T1-weighted images, in T2-weighted images, in DWI and in the related map of the apparent diffusion coefficient (ADC), presence or absence of contrast enhanced during the arterial, portal, and late phase in MRI, CT, and CEUS, and signal characteristic during the liver-specific phase in MRI of the ablation zone at 1, 3, and 6 months. IRE is a feasible, safe and efficient modality in the treatment of patients with non-resectable HCC. We had no major complication, even when the ablated lesion was adjacent to major branches of the portal vein. All images techniques showed similar accuracy during the follow-up at 1, 3, and 6 months in the assessment ablated zone.