Pre-treatment CEMRI habitat radiomics and deep learning for noninvasive prediction of the VETC pattern in hepatocellular carcinoma: an exploratory radiogenomic analysis - Takeaways - MDSpire

Pre-treatment CEMRI habitat radiomics and deep learning for noninvasive prediction of the VETC pattern in hepatocellular carcinoma: an exploratory radiogenomic analysis

  • By

  • Xingwu Xie

  • Yanxi Xiong

  • Xiao-Shan Huang

  • Xiaojuan Tang

  • Yue Zhao

  • Xiaoyu Xiao

  • Long Jin

  • June 26, 2026

  • 0 min

Share

  • 1

    Vessels encapsulating tumor clusters (VETC) in hepatocellular carcinoma (HCC) are linked to aggressive tumor behavior and immune evasion.

  • 2

    The study evaluated pre-treatment contrast-enhanced MRI radiomics and deep learning models for noninvasive prediction of VETC in HCC.

  • 3

    The fusion model combining intratumoral heterogeneity and deep learning features achieved an AUC of 0.901, outperforming individual models.

  • 4

    Transcriptomic analysis revealed significant differences in immune-related pathways between high-risk and low-risk HCC patient groups.

  • 5

    The study highlights the potential of CEMRI-based models to enhance clinical decision-making in the assessment of VETC in HCC.

Original Source(s)

Pre-treatment CEMRI habitat radiomics and deep learning for noninvasive prediction of the VETC pattern in hepatocellular carcinoma: an exploratory radiogenomic analysis

  • by Xingwu Xie, Yanxi Xiong, Xiao-Shan Huang, Xiaojuan Tang, Yue Zhao, Xiaoyu Xiao, Long Jin

  • June 26, 2026

Related Content

ESMO Breast Cancer 2026 Highlights: SERENA-6

Harold Burstein, MD, PhD, and Erica Mayer, MD, MPH discuss results from the SERENA-6 trial, which were presented at the 2026 ESMO Breast Cancer Congress.

  • June 26, 2026

  • 1 min
Correction: Follicular dendritic cell sarcoma associated with IgG4-related disease: a case report and literature review

The impact of the large B-cell lymphoma tumor microenvironment on efficacy of CD19-directed CAR-T immunotherapy