The purpose of this study was to evaluate the necessity of in-phase and opposed-phase MR images and their correlations with weight, the aspartate aminotransferase/alanine aminotransferase (AST/ALT) value, and age. Magnetic resonance spectroscopy (MRS) was used as a reference in this study. We selected 68 people as subjects, among which 14 were volunteers with normal AST/ALT values (<40/35 U/L) on a liver function study and 54 were non-alcoholic fatty liver patients for whom ultrasonic images had been obtained within 3 months of the study. In this study, the liver was more enhanced than the spleen or kidney. When the Eq. (3) formula was applied to normal volunteers, the difference between the in-phase and the opposed-phase images was −3.54 ± 12.56. The MRS study result showed a high sensitivity of 96.6% and a specificity of 100% (p = 0.000) when the cutoff value was 20%. Furthermore, this result showed a high sensitivity of 94% and a specificity of 80% with a similar cutoff when the Eq. (2) formula was applied to non-alcoholic fatty liver patients (p = 0.000). The MRS study revealed a strong correlation between normal volunteers and non-alcoholic fatty liver patients (r = 0.59, p = 0.04). The correlations between AST/ALT and Eq. (3) (r = 0.45, p = 0.004), age and Eq. (3) (r = 0.73, p = 0.03), and weight and Eq. (3) (r = 0.77, p = 0.000) values were all statistically significant. In the case of non-alcoholic liver disease, MRS was found to be significantly correlated with Eq. (1) (r = 0.39, p = 0.002), Eq. (2) (r = 0.68, p = 0.04), Eq. (3) (r = 0.67, p = 0.04), and AST/ALT (r = 0.77, p = 0.000). In conclusion, in-phase and opposed-phase images can help to distinguish a normal liver from a fatty liver in order to identify non-alcoholic fatty liver patients. The intensity difference between the in-phase and opposed-phase MR signals showed valuable correlations with respect to weight, AST/ALT value, and age, with all values being above the mild lipid value (r = 0.3).