Differentiation of adrenal adenomas from nonadenomas using CT histogram analysis method: A prospective study

Characterization of Adrenal Lesions on Unenhanced MRI Using Texture Analysis: A Machine‐Learning Approach

“…First‐order quantitative parameters represent the distribution of pixel intensities in an image (histogram analysis, HA); second‐order textural features are extracted by the analysis of gray‐level co‐occurrence matrices (GLCM) and rely on the relation between couples of pixels; finally, high‐order textural parameters extracted from neighborhood gray‐tone difference matrices (NGTDM) describe the relation between a pixel and the neighboring pixels, while run length matrices (RLNM) comprise the number of consecutive pixels that have the same intensity level and which occur in a specified direction. HA was first employed on both unenhanced and enhanced CT images for the diagnosis of adrenal adenomas by revealing the presence of negative pixels within the lesions suggestive of intralesional lipids; conversely, other authors reported the presence of negative pixels also within NAL, ie, metastasis, pheochromocytomas, and adrenal carcinoma . According to Blake et al, the most practical clinical application of HA seems to be supplementing unenhanced CT to improve sensitivity to almost 90%, maintaining a high specificity .…”

“…HA was first employed on both unenhanced and enhanced CT images for the diagnosis of adrenal adenomas by revealing the presence of negative pixels within the lesions suggestive of intralesional lipids 4,5 ; conversely, other authors reported the presence of negative pixels also within NAL, ie, metastasis, pheochromocytomas, and adrenal carcinoma. 6,7 According to Blake et al, the most practical clinical application of HA seems to be supplementing unenhanced CT to improve sensitivity to almost 90%, maintaining a high specificity. 8 To the best of our knowledge, only one study analyzed the role of HA applied to MR images and specifically on apparent diffusion coefficient (ADC) maps to differentiate AA from pheochromocytomas.…”

Differentiation between adrenal adenomas and nonadenomas using dynamic contrast-enhanced computed tomography

“…Many studies have confirmed the usefulness of attenuation measurement on unenhanced and delayed contrast-enhanced CT scan in the differentiation of adenomas from nonadenomas 7,10,18,19. All lesions of 10 HU or less on unenhanced CT images are lipid-rich adenomas 7.…”

Update on imaging of the adrenal cortex

“…The amount of lipid in the mass is proportional to the number of pixels with attenuation values of less than 0 Hounsfield units (negative pixels) within it. A study by Halefoglu et al [3] showed that 100% of adenomas contained negative pixels. A 90.9% consisted of more than 10% negative pixels.…”