Based on dynamic enhanced magnetic resonance imaging (MRI)
to explore the consistency of radiomics parameters obtained by different observers manually delineating region of interest (ROI).
Methods:
A total of 62 patients with breast cancer diagnosed in Fudan University Shanghai Cancer Center were selected. The ITK-SNAP software was used to the three- dimensional lesion delineation. The morphological characteristic and first-order parameters in the ROI
were extracted
and then the consistency test was carried out.
Results:
In the overall tumors
there were a higher inter-observer and intra-observer agreement (>0.75)
and the intra-observer agreement (all >0.9) was higher than inter-observers (all >0.8). When we divided tumors into mass-like and nonmass-like enhancement according to the tumor enhancement morphology
we found that the consistency of the morphological parameters was well (all >0.75)
but the consistency of peak and minimum values of the first-order parameter were less than 0.75. Among tumors of different T stages
the consistency of the first-order parameter peak and skewness of T
1
+T
2
tumors are less than 0.75
and the value of T
1
+T
2
tumors was generally lower than that of T
3
tumors.
Conclusion:
The consistency of the radiomics parameters between different observers is relatively high
which is a further confirmation of the reliability of the results of previous related studies. The consistency is still not appreciable among tumors with different T stages and different enhancement forms among some first-order parameters. Therefore
the sample size should be expanded for further stratified analysis.
Prediction of sentinel lymph node metastasis in breast cancer using multiparametric MRI radiomics and machine learning models
The value of radiomics features derived from the T2WI-FS assisted preoperative prediction of axillary lymph node metastasis of breast cancer
Radiomics in Breast Cancer
MRI radiomics in breast cancer
Value of a multiparameter MRI radiomics nomogram for preoperative prediction of endometrial carcinoma risk stratification
Related Author
Hongkai YANG
Xuan QI
Wuling WANG
Weiqun CHENG
Dong QI
Yongsheng HE
WANG Meng
LIU Zhou
Related Institution
Department of Radiology, Ma’anshan People’s Hospital, Ma’anshan
The Graduate School, Anhui Medical University
The Fifth Clinical Medical College of Anhui Medical University
Department of Diagnostic Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College