Deep multimodal state-space fusion of endoscopic-radiomic and clinical data for survival prediction in colorectal cancer - Takeaways - MDSpire

Deep multimodal state-space fusion of endoscopic-radiomic and clinical data for survival prediction in colorectal cancer

  • By

  • Ning Wang

  • Jiajing Lin

  • Wujin Li

  • Yahui Lyu

  • Yiqing Jiang

  • Zhizhan Ni

  • Qi Huang

  • Hong Chen

  • Qiang Yan

  • Chenshen Huang

  • December 31, 2025

  • 0 min

Share

  • 1

    HydraMamba is a multimodal framework that integrates endoscopic and CT data for improved lesion segmentation and survival prediction in colorectal cancer.

  • 2

    The model achieved state-of-the-art performance in lesion analysis, with Dice scores of 0.856 for endoscopy and 0.812 for CT imaging.

  • 3

    HydraMamba demonstrated calibrated survival modeling on the CT dataset, achieving a Harrell’s C index of 0.832 and an integrated Brier score of 0.161.

  • 4

    Combining endoscopic and CT information enhances diagnostic accuracy and prognostic modeling, addressing limitations of analyzing these modalities separately.

  • 5

    The study emphasizes the potential of multimodal learning in oncology to improve patient outcomes through advanced imaging and artificial intelligence.

Original Source(s)

Deep multimodal state-space fusion of endoscopic-radiomic and clinical data for survival prediction in colorectal cancer

  • by Ning Wang, Jiajing Lin, Wujin Li, Yahui Lyu, Yiqing Jiang, Zhizhan Ni, Qi Huang, Hong Chen, Qiang Yan, Chenshen Huang

  • December 31, 2025

Related Content

A generalizable 3D framework and model for self-supervised learning in medical imaging

An end-to-end deep learning pipeline for hematoma expansion prediction in spontaneous intracerebral hemorrhage based on non-contrast computed tomography

Deep learning model for assessing survival benefits in hepatocellular carcinoma patients undergoing intra-arterial therapies based on proliferative subtype