Background: Human amniotic epithelial cells (h-AECs), which have various merits as a cell source for cell therapy, are known to differentiate into cardiomyocytes in vitro. However, the ability of h-AECs to differentiate into cardiomyocytes in vivo, and their cell transplantation effects on myocardial infarction, are still unknown. In this study, we assessed whether h-AECs could differentiate into cardiomyocytes in vivo and whether h-AECs transplantation can decrease infarct size and improve cardiac function, in comparison to transplantation of cord blood-derived mesenchymal stem cells (MSCs) or adipose tissue-derived MSCs. Methods and Results: For our study, we injected h-AECs, cord blood-derived MSCs, adipose tissue-derived MSCs, and saline into areas of myocardial infarction in athymic nude rats. After 4 weeks, 3% of the surviving h-AECs expressed myosin heavy chain, a marker specific to the myocardium. Compared with the saline group, all cell-implanted groups showed a higher ejection fraction, lower infarct area by positron emission tomography and histology, and more abundant myocardial gene and protein expression in the infarct area. Conclusion: We showed that h-AECs can differentiate into cardiomyocyte-like cells, decrease infarct size, and improve cardiac function in vivo. The beneficial effects of h-AECs were comparable to those of cord blood and adipose tissue-derived MSCs. These results support the need for further studies of h-AECs as a cell source for myocardial regeneration due to their plentiful availability, low immunity and lack of ethical issues related to their use.