Objective To investigate the association between abdominal visceral fat content and the occurrence of type 2 diabetes mellitus （T2DM） and the potential of predicting the incidence of T2DM with abdominal visceral fat content. Methods The study included 45 patients with T2DM who underwent 3.0T MRI and 80 control patients， recruited from the First Affiliated Hospital of Baotou Medical College， Inner Mongolia University of Science and Technology. IDEAL software was utilized for automated image recombination to obtain fat fraction， relaxation rate （R2^*）， lipid phase， water phase， and in-phase images. Fat fraction images were transferred to AW 4.6 workstation for measurement of fat fraction plots by two observers using a double-blind approach. ITK-SNAP software was employed for semi-automatic delineation of abdominal visceral fat and subcutaneous adipose tissue through threshold segmentation method， followed by volume calculation. Statistical analysis of the data was performed using SPSS 26.0 software. Results （1） The T2DM group exhibited significantly higher levels of abdominal visceral adipocyte volume （VATV）， hepatic fat fraction （HFF）， pancreatic fat fraction （PFF）， triglyceride （TG）， and fasting blood glucose （FBG） compared to the control group. Conversely， the high density lipoprotein （HDL） level was significantly lower in the T2DM group than in the control group （P < 0.05）. The correlation analysis revealed that in male patients， abdominal VATV exhibited positive associations with body mass index （BMI）， waist circumference， subcutaneous adipose tissue volume （SATV）， hepatic fat fraction （HFF）， TG， and FBG； while displaying a negative association with HDL. In female patients， VATV demonstrated positive correlations with BMI， waist circumference， SATV， and HFF. （2） The ROC curve analysis demonstrated that the area under the curve （AUC） for PFF in diagnosing T2DM was 0.656 （P < 0.01，95%CI：0.534 ~0.777）. The determined cut-off value was 8.44%， yielding a sensitivity of 50% and specificity of 82.5%. Similarly， HFF exhibited an AUC of 0.744 （P < 0.05， 95%CI： 0.637 ~ 0.851）， with a cut-off value of 2.99%， sensitivity of 97.5%， and specificity of 42%. In male patients， VATV displayed an optimal threshold for T2DM diagnosis at a volume of approximately 3 466 cm3， resulting in a sensitivity of 63.2%， specificity of 89.5%， and AUC of 0.78 （P < 0.01）. The sensitivity was 78.9% and specificity was 84.2% when SATV， PFF and HFF were combined and the AUC was 0.839 （P < 0.01）. The optimal threshold for diagnosing T2DM with VATV in female patients was 2 103cm3， and the sensitivity， specificity and AUC of T2DM with VATV were 66.7%， 66.7% and 0.68 （P < 0.05）. When SATV， PFF and HFF were combined， the sensitivity was 90.5%. The specificity was 76.2% and the AUC was 0.909 （P < 0.01）. （3）For each additional unit of TG and FBG， the risk of T2DM respectively increased by 3986.3% and 417.1%， in other words， TG and FBG were independent predictors of the occurrence of T2DM. Conclusion The predictive value of pancreatic fat fraction， liver fat fraction， and abdominal visceral fat tissue volume for the onset of type 2 diabetes is limited. However， combining abdominal visceral adipose tissue volume with subcutaneous adipose tissue volume， pancreatic fat fraction， and liver fat fraction significantly enhances the efficacy， sensitivity， and specificity of predicting type 2 diabetes occurrence. The IDEAL-IQ technique offers a convenient method for quantitatively evaluating intra-abdominal ectopic adipose tissue deposition in patients with type 2 diabetes mellitus. It demonstrates advantages such as simplicity， good stability， high reproducibility and holds great potential.