Low-temperature calorimetry has been used to determine the ice content in concrete at different temperatures when exposed to low-temperature environments. However, the analysis of the ice content from the measured data of heat flow is not straightforward. In this study, two important factors influencing the ice content calculation are discussed. The importance of the baseline determination for the calculation of the ice content is realized. Two different methods of generating the baseline are discussed. First, the ‘J-baseline’ is discussed which is a recently proposed extrapolation method based on the accumulated heat curves measured in the freezing and the melting process. Second, the ‘C-baseline’ is discussed in which a calculated baseline is used where the heat capacity of both water and ice and the phase changing behaviour under different testing temperatures are considered. It turns out that both the ‘J-baseline’ method and the ‘C-baseline’ method can be used to calculate the approximate baseline. The heat of fusion of the water confined in small pores is another important parameter to be considered in ice content calculation. This property must be carefully analyzed in order to accurately calculate the ice contents at different temperatures in the freezing and melting process. It should be noted that there is no general agreement on how to obtain the important temperature dependence of the heat of fusion of water confined in small pores. By performing comparison studies, the present study shows the influence of the different values of the heat of fusion commonly adopted on the calculated ice content for the studied concrete samples. The importance and necessity to use an accurate value of the heat of fusion is emphasized. Based on the calculation of the baseline proposed in this work and by carefully selecting the values for the heat of fusion, the ice content in a hardened concrete sample is expected to be estimated with an acceptable accuracy.