Abstract

ObjectiveDifferentiation between renal oncocytoma (RON) and chromophobe renal cell carcinoma (chRCC) remains challenging. We aimed to assess the accurate apparent diffusion coefficient (ADC) radiomics features in differentiating these tumors. Materials and methodsThis single-center retrospective study included 14 patients with histopathologically proven RON (n=6) and chRCC (n=8) who underwent magnetic resonance imaging. Features were extracted from ADC maps. Features with an intraclass correlation coefficient >0.90, an intergroup p<0.01 and interrater differences with normal distribution underwent agreement and receiver operating characteristic curve analyses. ResultsOverall, 6 features qualified for further analysis and Bland-Altman plots revealed acceptable agreement for all. Only 1 first order feature and 5 high order texture features successfully predicted RON with more than 90% sensitivities and specificities more than 80%. ConclusionSquared mean ADC and certain gray level run length matrix features extracted by radiomics of ADC mapping provide quite high diagnostic precision in terms of distinguishing between RON and chRCC.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call