Comparative study of AI-assisted opportunistic CT and dual-energy X-ray bone mineral density detection in bone mass assessment of type 2 diabetes and non-diabetes patients
Objective To compare the diagnostic value of artificial intelligence(AI)-assisted opportun-istic CT and dual-energy X-ray absorptiometry(DXA)in the bone mass assessment of diabetic patients.Meth-ods The retrospective study method was adopted.A total of 72 patients receiving CT examination(such as abdomen,chest or spine)combined with lumbar vertebra DXA examination,diabetes screening or treatment in the First Affiliated Hospital of Army Military Medical University from January 1,2017 to May 31,2024 were selected as the study subjects and divided into the type 2 diabetic group(n=36)and non-diabetic group(n=36)according to the illness.The DXA scanning range was L1-L4 lumbar vertebral body;the AI quantitative CT-assisted bone density detection system imported the CT images,the interested region in the center of the T12-L2 vertebral body was labeled,and the bone density of cancellous bone in the vertebral body was auto-matically calculated.The diagnostic differences in osteopenia and osteoporosis were compared between the two modalities in the patients of different genders and ages in type 2 diabetes group and non-diabetes group.The receiver operating characteristic(ROC)curve was drawn to evaluate the specificity and sensitivity of the AI quantitative CT-assisted bone density detection system in the diagnosis of osteoporosis.Results There were no statistically significant differences in the detection rates of osteoporosis,osteopenia and normal bone density between the two methods(P＞0.05).The osteoporosis detection rate of DXA in the non-diabetic group and population＜60 years old was higher than that of the AI-quantitative CT-assisted bone mineral density detec-tion system,and the difference was statistically significant(P＜0.05).In the patients with different genders,there was no statistically significant difference in osteoporosis,osteopenia and normal bone density between the two methods(P＞0.05).Taking the DXA detection as the"gold standard",the AI quantitative CT-assis-ted bone mineral density detection system had a certain diagnostic value for osteoporosis[area under the curve(AUC)=0.661,95％CI:0.526-0.795,P=0.026],and its sensitivity and specificity were 68.1％and 64.0％re-spectively.Conclusion In the patients aged above 60 years old and with type 2 diabetes mellitus,the AI-assisted opportunistic CT has the same advantages as DXA in early detection of the patients at high risk of fractures,without increasing additional costs and unnecessary radiation risks.
type 2 diabetesosteoporosisartificial intelligencequantitative CTdual-energy X-ray bone mineral density test
万方数据
维普
