Gender-specific Data-Driven Adiposity Subtypes Using Deep-Learning-based Abdominal CT Segmentation.

Xiantong Zou,Xianghai Zhou,Yufeng Li,Qi Huang,Yuan Ni,Ruiming Zhang,Fang Zhang,Xin Wen,Jiayu Cheng,Yanping Yuan,Yue Yu,Chengcheng Guo,Guotong Xie,Linong Ji
DOI: https://doi.org/10.1002/oby.23741
IF: 9.298
2023-01-01
Obesity
Abstract:ObjectiveThe aim of this study was to quantify abdominal adiposity and generate data-driven adiposity subtypes with different diabetes risks. MethodsA total of 3817 participants from the Pinggu Metabolic Disease Study were recruited. A deep-learning-based recognition model on abdominal computed tomography (CT) images (A-CT model) was developed and validated in 100 randomly selected cases. The volumes and proportions of subcutaneous fat, visceral fat, liver fat, and muscle fat were automatically recognized in all cases. K-means clustering was used to identify subgroups using the proportions of the four fat components. ResultsThe Dice indices among the measurements assessed by the A-CT model and manual evaluation to detect liver fat, muscle fat, and subcutaneous fat areas were 0.96, 0.95, and 0.92, respectively. Three subtypes were generated separately in men and women: visceral fat dominant type (VFD); subcutaneous fat dominant type (SFD); and intermuscular fat dominant type (MFD). Compared with the SFD group, the MFD group had similar diabetes risk, and the VFD group had a 60% higher diabetes risk when age and BMI were adjusted for in men. The adjusted odds ratio for diabetes was 1.92 (95% CI: 1.32-2.78) in the MFD group and 6.14 (95% CI: 4.18-9.03) in the VFD group in women. ConclusionsThis study identified gender-specific abdominal adiposity subgroups, which may help clinicians to distinguish diabetes risk quickly and automatically.
What problem does this paper attempt to address?