Journal of Oncology / 2019 / Article / Tab 2 / Research Article
Texture Analysis of Dynamic Contrast-Enhanced MRI in Evaluating Pathologic Complete Response (pCR) of Mass-Like Breast Cancer after Neoadjuvant Therapy Table 2 Interobserver agreement for ROI drawing (Cronbach’s test).
Features Cronbach’s alpha 95% confidence interval F valuePost‐NAT enhancing volume 0.983 0.963, 0.992 57.30 <0.001 Interquartile range 0.965 0.927, 0.983 28.78 <0.001 Skewness 0.410 −0.240, 0.719 1.70 0.081 75 percentile 0.972 0.941, 0.987 35.40 <0.001 Uniformity 0.815 0.612, 0.912 5.41 <0.001 Median 0.938 0.870, 0.971 16.21 <0.001 Energy 0.997 0.995, 0.999 396.06 <0.001 Robust mean absolute deviation 0.969 0.935, 0.985 32.14 <0.001 Mean absolute deviation 0.972 0.940, 0.986 35.14 <0.001 Maximum 0.993 0.985, 0.997 142.62 <0.001 Root mean squared 0.959 0.913, 0.980 24.28 <0.001 90 percentile 0.982 0.961, 0.991 54.52 <0.001 Minimum −0.140 −1.395, 0.457 0.88 0.637 Entropy 0.972 0.941, 0.987 35.57 <0.001 Range 0.989 0.976, 0.995 88.52 <0.001 Variance 0.967 0.931, 0.984 30.66 <0.001 10 percentile 0.725 0.422, 0.869 3.64 <0.001 Kurtosis 0.877 0.743, 0.942 8.16 <0.001 25 percentile 0.857 0.700, 0.932 7.01 <0.001 Mean 0.944 0.883, 0.974 18.02 <0.001
Features with Cronbach’s alpha >0.9.