藤井 康友¹, 谷口 信行¹, 王 怡¹, 重田 浩一朗¹, 尾本 きよか¹, 伊東 紘一¹, 曹 景文², 隈崎 健二², 伊藤 貴司², 高山 智司²

Yasutomo FUJII¹, Nobuyuki TANIGUCHI¹, Yi WANG¹, Kouichiro SHIGETA¹, Kiyota OMOTO¹, Kouichi ITOH¹, Jing-Wen TSAO², Kenji KUMAZAKI², Takashi ITOH², Tomotsugu TAKAYAMA²

¹自治医科大学臨床検査医学, ²アロカ(株)

¹Department of Clinical Laboratory Medicine, Jichi Medical School, ²Aloka Co. Ltd

キーワード : Histogram, Liver cirrhosis, Radio-frequency signal, Tissue characterization, Ultrasound

We used texture analysis in conjunction with an alternative method of analyzing the amplitude histogram by using a radio-frequency (RF) signal to differentiate ultrasonograms from normal and cirrhotic livers. This method segments the region of interest (ROI) into multiple layers (sub-ROIs). In each sub-ROI of a homogenous medium, the histogram of envelopedamplitude of RF backscattered echoes resembles a Rayleigh distribution. Theoretically, the values of signal-noise ratio (SNR), skewness, and kurtosis for Rayleighstatistics are constant and independent of mean scattering intensity, whichis contributed by such undesirable effects as tissue attenuation, beam diffraction, and incident wave forms. Thus these values, which averaged overall sub-ROI, should provide an unbiased estimator. Thirty-six cases of normal liver and 28 cases of cirrhotic liver, all confirmed by clinical findings including laboratory and pathological data, were studied, and SNR, skewness, and kurtosis values of the disease groups were compared. At the same time, in the conventional method, which did not segment the ROI into multiple sub-ROIs, these values were estimated. The unpaired t-test was used to determine statistical significance. With the new method, all values obtained from cirrhotic livers differed significantly from those obtained from normal livers, and the standard deviation of these values was smaller than those obtained using the conventional method. These results thus suggest that the new method can be used to diagnose cirrhotic liver, objectively.