Click here to


Are you sure ?

Yes, do it No, cancel

Diagnostic Value of 0.35T MR Radiomic Features in Recognition of Tumor Origin in Liver Cancer

G Simpson*, D Asher, J Ford, L Portelance, F Yang, N Dogan, University of Miami, Miami, FL


(Sunday, 7/29/2018) 3:00 PM - 6:00 PM

Room: Exhibit Hall

Purpose: MR-guided radiotherapy machines are becoming popular in radiation oncology for their ability to accurately position patients without added dose. Low field strength (0.35T) images provide superior soft tissue contrast compared to conventional onboard imaging (CBCT). The objective of the current study is to examine the capability of pretreatment image radiomic texture features to differentiate primary vs metastatic liver disease.

Methods: Five patients undergoing SBRT treatment of liver lesions were included in the study. Three were diagnosed with secondary malignant neoplasm of the liver and intrahepatic bile duct and two with hepatocellular carcinoma. Gray-level size zone matrix (GLSZM) based texture features were extracted from physician-defined contours of the gross tumor volume (GTV) to differentiate tumor origin. Wilcoxon signed-rank test (JMP Pro, SAS Institute, Cary, NC) was employed for comparison of texture features between the groups using a significance level of p-value < 0.05.

Results: A simple univariate analysis identified seven of the 13 GLSZM features as statiscally significant, despite a small patient population. The features found to significantly differentiate lesion origin are short zone emphasis (SZE), large zone emphasis (LZE), zone size non-uniformity (ZSN), zone percentage (ZP), low gray-level zone emphasis (LGLZE), short zone low gray-level emphasis (SZLGLE), and large zone high gray-level emphasis (LZHGLE). The highest percent difference seen between the medians of each group was observed in LZHGLE (13.2%), and the most similar was SZE (0.8%).

Conclusion: These preliminary results indicate that for patients ineligible for biopsies, lesion origin may be obtainable from pretreatment low field images. Differentiation of primary from secondary malignancies based on imaging may provide supplementary knowledge and insight for treatment modification for patients unable to undergo biopsies.


Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email