The imbalances between the In-phase (I) and Quadrature-phase (Q) branches represent a significant source of impairment in the orthogonal frequency division multiplexing (OFDM) systems. Recently, it has been shown that the unwanted IQ imbalances can be actually exploited to achieve a diversity gain. In this contribution, by taking into account the diversity gain resulting from the IQ imbalances, we develop a novel algorithm to jointly estimate the channel impulse response and IQ imbalances occurring at both the transmitter and receiver. Starting from the Maximum Likelihood (ML) principle, we derive an estimation algorithm based on the expectation maximization (EM) algorithm, which exploits information from the pilot symbols and detected data symbols in a systematic fashion. To reduce the complexity of the estimation algorithm, a sub-optimal scheme is also introduced. The results indicate that the proposed algorithms achieve a significant improvement in the bit error rate (BER) performance after three iterations as compared to conventional data-aided algorithms.