Abstract
Purpose: Understanding how the pancreas changes is critical for detectingdeviations in type 2 diabetes and other pancreatic disease. We measure pancreassize and shape using morphological measurements from ages 0 to 90. Our goalsare to 1) identify reliable clinical imaging modalities for AI-based pancreasmeasurement, 2) establish normative morphological aging trends, and 3) detectpotential deviations in type 2 diabetes. Approach: We analyzed a clinically acquired dataset of 2533 patients imagedwith abdominal CT or MRI. We resampled the scans to 3mm isotropic resolution,segmented the pancreas using automated methods, and extracted 13 morphologicalpancreas features across the lifespan. First, we assessed CT and MRImeasurements to determine which modalities provide consistent lifespan trends.Second, we characterized distributions of normative morphological patternsstratified by age group and sex. Third, we used GAMLSS regression to modelpancreas morphology trends in 1350 patients matched for age, sex, and type 2diabetes status to identify any deviations from normative aging associated withtype 2 diabetes. Results: When adjusting for confounders, the aging trends for 10 of 13morphological features were significantly different between patients with type2 diabetes and non-diabetic controls (p < 0.05 after multiple comparisonscorrections). Additionally, MRI appeared to yield different pancreasmeasurements than CT using our AI-based method. Conclusions: We provide lifespan trends demonstrating that the size and shapeof the pancreas is altered in type 2 diabetes using 675 control patients and675 diabetes patients. Moreover, our findings reinforce that the pancreas issmaller in type 2 diabetes. Additionally, we contribute a reference of lifespanpancreas morphology from a large cohort of non-diabetic control patients in aclinical setting.