Objective To explore the diagnostic value of 1H proton MRS and in-phase and out-phase MRI for bone marrow fat deposition (BMFD), and to compare the consistency of two methods. Methods Thirsty-eight patients who were diagnosed with osteoporosis using DXA and 30 people with normal bone mass were enrolled in this study. All the 68 subjects underwent MRI regular examination, MRS, and in-phase and out-phase MRI. SIR, LRW, and FF of L2 vertebra were measured. SIR, LRW, and FF between the two groups were compared using Student t test. Diagnostic accuracy of SIR, LRW, and FF for osteoporosis were analyzed using ROC curve, and the diagnostic threshold of these indicators for osteoporosis was determined. By the use of Kolmogorov-Snimov normality test, the data were proved to be normally distributed. The correlations among SIR, LRW, and FF were assessed with Pearson analysis. Results Sixty-eight axial and sagittal MRI in-phase and out-phase MR images of the lumbar spine and 68 single voxel MRS images were obtained. SIR, LWR, and FF in osteoporosis group were 0.498 ±0.160, 3.677 ± 3.093, and 0.732 ±0.176, respectively. SIR, LWR, and FF in normal bone mass group were 0.350 ±0.971, 2.094 ±1.892, and 0.573 ±0.211, respectively. There were statistically significant differences in SIR, LWR, and FF between the two groups (t values were 6.404, 2.035, and 2.738 respectively, P <0.05). The area (measured with the SIR deferential point) under ROC curve was 0.784. Using the SIR value of 0.3850 as the ROC cut off point, the diagnostic sensitivity and specificity were 78.9% and 66.7%, respectively. In MRS, using the value 2.063 of LWR and 0.674 of FF as the cutting off points, the area under ROC curve was 0.706 and 0.740, the diagnostic sensitivity was 75.0% and 79.2%, and the diagnostic specificity was 71.4% and 72.4%, respectively. The result of correlation analysis between SIR, LRW, and FF using Pearson correlation showed P <0.05. Conclusion Both 1H-MRS and in-phase and out-phase MRI can provide information of bone marrow fat deposition in the lumbar spine and help to achieve quick and non-invasive osteoporosis diagnosis. There is correlation between the indicators of the two methods. [ABSTRACT FROM AUTHOR]