Research Article

Ability of 18F-FDG PET/CT Radiomic Features to Distinguish Breast Carcinoma from Breast Lymphoma

Table 2

The results of ROC analysis of potentially optimal texture parameters in PET and CT images for breast lymphoma versus breast carcinoma.

VariableLymphomaCarcinomaAUCP value
MedianRangeMedianRange

Radiomic parameters on PET
 HISTO_entropy_log101.360.64–1.760.970.32–1.660.7400.001
 SHAPE_volume(# vx)183.0022.00–6334.0071.0011.00–4241.000.7250.002
 GLCM_entropy_log102.041.00–3.191.540.59–3.030.7300.001
 GLRLM_HGRE287.9517.14–1919.1575.669.96–862.280.752<0.001
 NGLDM_Contrast0.330.04–2.320.160.00–0.590.7120.003
 GLZLM_HGZE266.6217.53–1593.6483.779.67–832.220.754<0.001

Radiomic parameters on CT
 HISTO_Kurtosis5.021.90–217.172.841.49–97.340.6600.024
 SHAPE_volume (mL)11.381.07–403.074.400.73–272.410.7280.001
 GLCM_Homogeneity0.490.08–0.600.440.33–0.650.6280.072
 GLRLM_RLNU977.06135.49–24346.21507.3580.31–19527.320.6850.009
 NGLDM_Busyness0.530.04–6.040.380.13–2.770.6020.152
 GLZLM_LZHGE14709543.6377721.12–1603305720.991385779.2244958.07–435410392.460.6810.011

Abbreviations: HISTO, histogram; GLCM, gray-level co-occurrence matrix; GLRLM, gray-level run-length matrix; HGRE, high gray-level run emphasis; GLZLM, gray-level zone-length matrix; HGZE, high gray-level zone emphasis; RLNU, run-length nonuniformity; LZHGE, long-zone high gray-level emphasis.