The ability and net benef it of attenuation measurement and histogram analysis for the dif ferentiating minimal- fat renal angiomyolipoma from chromophobe/papillary renal cell carcinoma on unenhanced CT
|更新时间:2025-12-15
|
The ability and net benef it of attenuation measurement and histogram analysis for the dif ferentiating minimal- fat renal angiomyolipoma from chromophobe/papillary renal cell carcinoma on unenhanced CT
The ability and net benef it of attenuation measurement and histogram analysis for the dif ferentiating minimal- fat renal angiomyolipoma from chromophobe/papillary renal cell carcinoma on unenhanced CT
e the ability of mean attenuation and histogram analysis for the differentiating minimal-fat renal angiomyolipoma (AML) from chromophobe/papillary renal cell carcinoma (RCC) and their net benef it on unenhanced CT.
Methods:
Unenhanced CT images of 37 minimal-fat AML (41 lesions)
32 chromophobe and 30 papillary RCC patients were analyzed
the mean attenuation and relative histogram parameters (10
th
value of attenuation) of the lesions were measured and calculated. The efficiency of the parameters for differentiating of minimal-fat AML from RCC were assessed by receiver operating characteristic (ROC) curve. Decision curve analysis (DCA) were performed to evaluate the net benef it of the two parameters.
Results:
The area under curve (AUC) of the parameters for the differentiating minimal-fat AML from papillary RCC were 0.83 and 0.85
respectively
and no signif icant difference was observed between them. There was neither signif icant difference between the AUC of the parameters for the differentiating minimal-fat AML from chromophobe RCC 0.71 and 0.78. In the differentiating ofminimal-fat AML and papillary RCC
when the diagnosis was applied basing on mean or 10
th
value of attenuation with a threshold probability (malignant probability) of higher than 30%
the net benef it was higher than treating all patients. And for chromophobe RCC and minimal-fat AML
when the diagnosis was applied basing on mean attenuation with a threshold probability (malignant probability) of higher than 35%
the net benef it was higher than treating all patients. When the diagnosis was applied basing on 10
th
value of attenuation
the net benef it was higher than treating all at any threshold probability. The net benef it of 10
th
value of attenuation was higher than mean attenuation at threshold probability less than 45%
but was lower at threshold probability of 45%~80%.
Conclusion:
Unenhanced CT attenuation measurement and histogram analysi
s improved the distinguishing minimal-fat AML from chromophobe/papillary RCC
and provides def inite and difference net benef it for patients.
CT/MRI features of renal carcinomas associated with Xp11.2 translocation/TFE3 gene fusions
Ability of attenuation measurement and histogram analysis in differentiating minimal-fat renal angiomyolipoma from papillary and chromophobe renal cell carcinoma on unenhanced CT
Clinical study of CTU simple score in differential diagnosis of renal parenchymal infiltrating renal pelvis carcinoma and renal pelvis infiltrating renal cell carcinoma
Application progress of deep learning techniques in lung cancer imaging diagnosis
Radiomics models in differentiating parotid pleomorphic adenoma from Warthin tumor
Related Author
WANG Lu
XU Pengju
ZHU Mingming
SHEN Lijuan
DING Zhu
XIAO Jing
ZHANG Dongxing
LIU Wenhui
Related Institution
Department of Radiology, Zhongshan Hospital, Fudan University
Department of Radiology, Taizhou Second People&rsquo
Department of Imaging, Huai&rsquo
Department of Nuclear Medicine, Dazhou Central Hospital
Department of Radiology, Fudan University Shanghai Cancer Center, Department of Oncology, Shanghai Medical College, Fudan University