衰减系数
衰减
脂肪变性
磁共振成像
相关系数
磁共振弹性成像
超声波
核磁共振
脂肪肝
线性回归
医学
弹性成像
核医学
内科学
放射科
物理
数学
统计
疾病
光学
作者
Toshifumi Tada,Takashi Kumada,Hidenori Toyoda,Satoshi Yasuda,Yasuhiro Sone,Shinichi Hashinokuchi,Sadanobu Ogawa,Takuma Oguri,Naohisa Kamiyama,Makoto Chuma,Tomoyuki Akita,Junko Tanaka
摘要
Recently, a new method has been developed to diagnose hepatic steatosis with attenuation coefficients based on the ultrasound-guided attenuation parameter (UGAP). We investigated whether fibrosis identified by hepatic stiffness measurements based on magnetic resonance elastography (MRE) affects attenuation coefficient measurement using UGAP for the evaluation of hepatic steatosis.A total of 608 patients with chronic liver disease were analyzed. Correlations between magnetic resonance imaging-determined proton density fat fraction (PDFF) or MRE value and attenuation coefficients were evaluated. In addition, the interaction between hepatic fibrosis and the attenuation coefficient was analyzed.The correlation coefficient (r) between PDFF values and attenuation coefficient values was 0.724, indicating a strong relationship. Conversely, the r between MRE values and attenuation coefficient values was -0.187, indicating almost no relationship. In the multiple regression assessment of the effect of PDFF and MRE on the attenuation coefficient based on UGAP, the P-values for PDFF, MRE, and PDFF × MRE were < 0.001, 0.277, and 0.903, respectively. In patients with non-alcoholic fatty liver disease (n = 169), the r between PDFF values and attenuation coefficient values was 0.695, indicating a moderate relationship. Conversely, the r between MRE values and attenuation coefficient values was -0.068, indicating almost no relationship. In the multiple regression assessment of the effect of PDFF and MRE on the attenuation coefficient based on UGAP, the P-values for PDFF, MRE, and PDFF × MRE were <0.001, 0.948, and 0.706, respectively.UGAP-determined attenuation coefficient was weakly affected by liver stiffness, an indicator of hepatic fibrosis.
科研通智能强力驱动
Strongly Powered by AbleSci AI