Introduction: This paper is aimed to evaluate the feasibility of computed tomography (CT) based thermometry for radiofrequency ablation (RFA) through the investigation on the effects of principal CT acquisition parameters to the CT number. Methods: The effects of CT acquisition parameters (tube voltage, tube current, gantry rotation time, and CT reconstructed slice thickness), as well as metal artefacts on CT number were evaluated by using liver tissue equivalent polyacrylamide (PAA) phantom. The correlation between CT number and tissue temperature from 37 0C to 80 0C was studied with the use of PAA phantom with optimum CT acquisition parameters. Results: The CT numbers show insignificant changes under variation of tube voltages, tube currents, gantry rotation time, and CT reconstructed slice thickness respectively. The CT number difference obtained for each series of the variations are less than 2 HU, which indicates the changes in the CT acquisition parameters has insignificant effects on the CT number shift. On the other hand, the CT reconstructed image is able to improve the metal artefact caused by RF electrode through the increases of CT reconstruction condition. In this paper. A linear regression analysis on the CT number and temperature suggested the CT number is inverse linearly proportional to temperature, with a CT thermal sensitivity of –0.521 ± 0.061 HU/0C (R2 = 0.998). Conclusion: In summary, the results show that assessment of CT thermometry is feasible with the use of current CT technology as it produces high reproducibility and stable CT measurement, which is proved to be independent for CT acquisition parameters.