Type 2 diabetes mellitus (T2DM) is defined by progressive pancreatic dysfunction, with histopathological studies revealing the gland as a primary site of injury. Key alterations—termed diabetic pancreatopathy—include exocrine atrophy, interlobular fibrosis, and fatty infiltration, often preceding clinical hyperglycemia [1,2].

Conventional imaging techniques such as computed tomography (CT) and ultrasonography (US) detect late-stage morphological changes in diabetic pancreatopathy, including reduced parenchymal attenuation and increased echogenicity reflective of fat accumulation [3,4]. However, these modalities cannot resolve microstructural or microvascular alterations critical to early disease progression. Quantitative MRI (QMRI) methods—notably proton density fat fraction (PDFF) mapping and dynamic contrast-enhanced MRI (DCE-MRI)—have enabled objective quantification of pancreatic steatosis, fibrosis, and macrovascular perfusion [[5], [6], [7]]. Despite their utility, current QMRI approaches suffer from two fundamental constraints: they compartmentalize structural (e.g., PDFF) and vascular (e.g., DCE-MRI) parameters into isolated assessments, obscuring microstructural-perfusion interactions; moreover, they treat the functionally heterogeneous pancreas as a uniform organ, disregarding documented craniocaudal gradients in islet density, blood flow, and metabolic vulnerability [8,9]. Although dual-energy CT studies suggest regional variations in fat distribution in type 2 diabetes [10,11], this modality cannot provide non-contrast, quantitative assessment of subregional microvascular perfusion and involves ionizing radiation.

Intravoxel incoherent motion (IVIM) MRI addresses these limitations by simultaneously decoding true molecular diffusion (D, reflecting cellular integrity and fibrosis) and perfusion fraction (F, representing microvascular flow) without contrast agents [12]. While IVIM has proven valuable in pancreatic oncology for differentiating adenocarcinoma and correlating with microvessel density [13,14], its potential to dissect spatially heterogeneous microenvironments in T2DM pancreatopathy is untapped. Crucially, no prior study has leveraged IVIM to map region-specific microstructural and microvascular pathology across the head, body, and tail of the diabetic pancreas.

This study aims to establish IVIM MRI as a quantitative biomarker of diabetic pancreatopathy by mapping region-specific microstructural and microvascular alterations, thereby enabling earlier detection and targeted interventions in T2DM.