Staging Liver Fibrosis with Hepatic Perivascular Adipose Tissue as a CT Biomarker

Skylar Chan, Tejas Sudharshan Mathai, Praveen T.S. Balamuralikrishna, Vivek Batheja, Jianfei Liu, Meghan G Lubner, Perry J Pickhardt, Ronald Summers

Keywords: CT, Liver Fibrosis, Cirrhosis, Perivascular Adipose Tissue, Portal Vein

TL;DR: Biomarkers of the fat or perivascular adipose tissue (PVAT) surrounding the vessels in the liver (portal vein and other hepatic vessels) can be used to stage fibrosis.

Abstract: Cirrhosis is the 12th leading cause of death in the US. There are several CT imaging signs of late fibrosis, such as redistribution of liver segment volume, increased liver nodularity, and periportal space widening. Timely intervention can reverse the progression of early hepatic fibrosis, but later stages are irreversible. We hypothesize that the perivascular adipose tissue (PVAT) around the portal vein arising from periportal space widening may also be predictive of liver fibrosis. In this work, a fully automated pipeline was developed to segment the liver, spleen, portal vein and its branches. The PVAT in the vicinity of the portal vein was identified. From these structures, CT imaging biomarkers (volume, attenuation, fat fraction) were computed. They were used to build uni- and multivariate logistic regression models for diagnosing advanced fibrosis and cirrhosis. The best multivariate model for cirrhosis achieved 93.3% AUC, 78.9% sensitivity, and 93.4% specificity. For advanced fibrosis, the multivariate model obtained 88.7% AUC, 84.2% sensitivity, and 73.7% specificity. The automated approach may be useful for population-based studies of metabolic disease and opportunistic screening.

Primary Subject Area: Segmentation

Secondary Subject Area: Detection and Diagnosis

Paper Type: Validation or Application

Registration Requirement: Yes

Visa & Travel: Yes

Midl Latex Submission Checklist: Ensure no LaTeX errors during compilation., Created a single midl25_NNN.zip file with midl25_NNN.tex, midl25_NNN.bib, all necessary figures and files., Includes \documentclass{midl}, \jmlryear{2025}, \jmlrworkshop, \jmlrvolume, \editors, and correct \bibliography command., Did not override options of the hyperref package, Did not use the times package., Author and institution details are de-anonymized where needed. All author names, affiliations, and paper title are correctly spelled and capitalized in the biography section., References must use the .bib file. Did not override the bibliographystyle defined in midl.cls. Did not use \begin{thebibliography} directly to insert references., Tables and figures do not overflow margins; avoid using \scalebox; used \resizebox when needed., Included all necessary figures and removed *unused* files in the zip archive., Removed special formatting, visual annotations, and highlights used during rebuttal., All special characters in the paper and .bib file use LaTeX commands (e.g., \'e for é)., Appendices and supplementary material are included in the same PDF after references., Main paper does not exceed 9 pages; acknowledgements, references, and appendix start on page 10 or later.

Submission Number: 43