Skip to main content
Fig. 1 | Cancer Imaging

Fig. 1

From: MRI-based radiomics to predict lipomatous soft tissue tumors malignancy: a pilot study

Fig. 1

Radiome extraction pipeline. Size and shape features were extracted from the binary mask. Intensity distribution features were extracted from masked MR images from the histogram built with 256 bins. Image gray levels were discretized in a smaller number of gray levels with an equal probability algorithm. Images were discretized in 8, 16, 24, 32, 40, 48, and 64 Gy levels. For each discretization level, four matrices were built: GLCM (Gray-level co-occurrence matrix), GLRLM (Gray-level run length matrix), GLSZM (Gray-level size zone matrix), and NGTDM (Neighborhood gray tone difference matrix) from which characteristics were extracted, then averaged. Frequency domain-based texture features were extracted using a Gabor filtering

Back to article page