Estimation of the aggregate interference is required in order to predict the performance of a wireless system in its working environment. Body Area Networks (BANs) for healthcare applications are becoming a reality, allowing patients to be monitored continuously without forcing them to stay in bed or in hospital. The increasing number of wireless medical devices makes the ISM (Industrial, Scientific and Medical) band particularly crowded. If new smart BANs have to correctly operate in hospital, coexistence with the existing wireless devices must be accurately investigated, starting with studying the interference in the operating frequency band. The intensity of the interference is strongly related to the chosen threshold, which defines if a received sample has to be categorized as interference or noise. Adaptive threshold methods outperform the non-adaptive ones, due to flexibility and robustness. Among the adaptive threshold methods, the forward consecutive mean excision (FCME) is one of the most attractive, since it is blind, computationally simple and efficient. When applied to large data set, it may require too long time to be computed, thus median filtering has been proposed (med-FCME). In this paper we compare the performance of fixed vs adaptive threshold methods. The two methods are applied to a set of real measurements taken in a modern city hospital over one week.