本文選題：癌　切入點：腎細胞　出處：《中國醫(yī)學影像技術(shù)》2017年12期

【摘要】：目的探討定量CT紋理分析鑒別透明細胞型腎癌和非透明細胞型腎癌的可行性。方法回顧性分析100個透明細胞型腎癌和27個非透明細胞型腎癌病灶的CT圖像,應用TexRAD軟件分析各掃描期相兩種類型腎癌的紋理特征。結(jié)果增強圖像上,非透明細胞型腎癌的平均灰度值、標準差、熵、正像素的平均值明顯低于透明細胞型腎癌,而峰度高于透明細胞型腎癌(P均0.001),偏度差異無統(tǒng)計學意義(P0.05)。在皮髓質(zhì)期的粗糙紋理上,正像素的平均值鑒別兩種類型腎癌的ROC曲線下面積為0.92±0.04,敏感度為0.85,特異度為0.93,準確率為0.87。結(jié)論透明細胞型腎癌與非透明細胞型腎癌的CT紋理特征間存在顯著差異,定量CT紋理分析鑒別診斷這兩種類型腎癌具有臨床價值。
[Abstract]:Objective to study the feasibility of quantitative CT texture analysis in differentiating clear cell renal cell carcinoma from non-transparent cell renal cell carcinoma.Methods CT images of 100 clear cell renal cell carcinoma and 27 opaque cell renal cell carcinoma were analyzed retrospectively. The texture features of two types of renal cell carcinoma were analyzed by TexRAD software.Results the average gray value, standard deviation, entropy and positive pixel of non-transparent cell renal cell carcinoma were significantly lower than those of transparent cell renal cell carcinoma, but the kurtosis was higher than that of transparent cell renal cell carcinoma (P = 0.001), and there was no significant difference in bias (P 0.05).On rough texture of medullary phase, the area under ROC curve of positive pixel was 0.92 鹵0.04, sensitivity was 0.85, specificity was 0.93and accuracy was 0.87.Conclusion there are significant differences in CT texture features between clear cell renal cell carcinoma and non-transparent cell renal cell carcinoma. Quantitative CT texture analysis is of clinical value in differential diagnosis of these two types of renal cell carcinoma.
【作者單位】： 中國醫(yī)學科學院北京協(xié)和醫(yī)學院北京協(xié)和醫(yī)院放射科;
【基金】：衛(wèi)生部行業(yè)基金科研專項(201402019)
【分類號】：R730.44;R737.11

【相似文獻】

相關(guān)期刊論文 前2條

1 王世民;成人T細胞型白血病的新檢驗法[J];日本醫(yī)學介紹;1985年06期

2 向美榮;;燕麥細胞型肺癌伴精神障礙的護理[J];山東醫(yī)藥;2005年31期

相關(guān)博士學位論文 前2條

1 李振江;基于MRI的多維放療技術(shù)若干關(guān)鍵問題的研究[D];東南大學;2017年

2 鐘紅霞;MRI紋理分析在預測直腸癌NCRT療效、TN分期、神經(jīng)脈管侵犯方面的作用[D];北京協(xié)和醫(yī)學院;2017年

相關(guān)碩士學位論文 前2條

1 虎曉軍;腎細胞癌的CT影像學表現(xiàn)與病理學類型的相關(guān)性研究[D];寧夏醫(yī)科大學;2017年

2 凌茜;DCE-MRI全定量參數(shù)及紋理分析應用于肝臟良惡性病變的診斷效能[D];重慶醫(yī)科大學;2017年

，

本文編號：1698745