The growing incidence of heart failure and the limitations of cardiac transplantation and totally implantable artificial hearts justify the search for alternative therapies. Selfreconstruction of the heart might be one of them. It is based on the use of an animal-derived decellularized scaffold reseeded with cells able to reconstitute a contractile, vascular and valvular pattern affording normal cardiac function. Whereas decellularization techniques are currently well controlled, the choice of cell type to generate the different constituents of cardiac tissue (cardiomyocytes, endothelial cells, smooth vascular cells, myofibroblasts), and the optimal mode of cell transfer, are still far from clear. Furthermore, scaffold recolonization is not only dependent on the phenotype of the grafted cells: it is also influenced by the nature of the biochemical signals emitted by the scaffold and by the physical state of the substrate. Modulation of these two factors can influence the contractile performance of the self-rebuilt organ. The complexity of these challenges is such that total replacement of the heart by a self-built organ is at best a far-off perspective. However, continued pursuit of this objective stimulates the development of bioengineering techniques, and partial replacement of heart tissue by cell-seeded scaffolds appears more feasible. Applications could include the strengthening of infarcted myocardium and, possibly, cardiac valve replacement.