The estimated prevalence of non-alcoholic fatty liver disease (NAFLD) worldwide is approximately 25%. However, the real prevalence of NAFLD and the associated disorders is unknown mainly because reliable and applicable diagnostic tests are lacking. This is further complicated by the lack of consensus on the terminology of different entities such as NAFLD or nonalcoholic steatohepatitis (NASH). Although assessing fatty infiltration in the liver is simple by ultrasound, the gold standard for the assessment of fibrosis, the only marker of progression towards more severe liver disease is still liver biopsy. Although other non-invasive tests have been proposed, they must still be validated in large series. Because NAFL/NAFLD/NASH and related metabolic diseases represent an economic burden, finding an inexpensive method to diagnose and stage fatty liver is a priority. A translational approach with the use of cell and/or animal models could help to reach this goal.