Meeting Banner
Abstract #1939

The value of diffusion kurtosis imaging texture analysis in predicting treatment response to transcatheter arterial chemoembolization in HCC

Chen Qiao1, Ailian Liu1, Ying Zhao1, Qingwei Song1, Xin Li 1, Yan Guo2, and Tingfan Wu2
1the First Affiliated Hospital of Dalian Medical University, DaLian, China, 2Philips Healthcare, BeiJing, China


This work aimed at texture analysis based on mean apparent diffusion coefficient (MD) of diffusion kurtosis imaging (DKI) to evaluate the treatment efficacy of transcatheter arterial chemoembolization (TACE) in hepatocellular carcinoma (HCC). The results showed that MD texture features can differentiate objective response and non-response groups. When using the single parameter, the GreyLevelNonuniformity_angle135_offset1 achieved the best result (AUC: 0.714, sensitivity: 68.2%, specificity: 81%). In the combined diagnosis of multiple parameters, FrequencySize,MaxIntensity, GreyLevelNonuniformity_angle0_offset1 and GreyLevelNonuniformity_angle135_offset1 were combined to obtain the best combined diagnosis(AUC:0.760,sensitivity: 81.8%, specificity: 66.7%).

This abstract and the presentation materials are available to members only; a login is required.

Join Here