Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jul 18, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

A flexible deep learning framework for liver tumor diagnosis using variable multi-phase contrast-enhanced CT scans.

Shixin Huang1,2, Xixi Nie3, Kexue Pu4

  • 1Department of Scientific Research, The People's Hospital of Yubei District of Chongqing city, Chongqing, 401120, China.

Journal of Cancer Research and Clinical Oncology
|October 3, 2024
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Focusing scattered light through live tissue with channel-selected wavefront shaping.

Biomedical optics express·2026
Same author

An electrochemiluminescence biosensor based on the hairpin-mediated exponential amplification and CRISPR/Cas12a amplification for ultrasensitive detection of MMP-2.

Colloids and surfaces. B, Biointerfaces·2026
Same author

Graph Neural Network-Assisted Machine Learning for High-Throughput Buried Interfacial Materials Screening in Antisolvent-Free Perovskite Solar Cells.

ACS applied materials & interfaces·2026
Same author

Factors Affecting the Evolution of Professional Identity from Nursing Student to Registered Nurse: A Longitudinal Trajectory Analysis.

Nurse educator·2026
Same author

Reducing P-O bond covalency in polyanionic groups to increase the intrinsic electronic conductivity in sodium positive electrodes.

Nature communications·2026
Same author

High-Performance Flexible Electromagnetic Wave Absorber through Synergistic Regulation of Matching and Reflective Layers.

ACS applied materials & interfaces·2026

A new Hierarchical Long Short-Term Memory (H-LSTM) model accurately distinguishes liver cancer types using contrast-enhanced CT (CECT) images. This AI-driven approach improves preoperative diagnosis, even with incomplete imaging phases.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Liver cancer is a leading cause of cancer mortality globally.
  • Accurate preoperative diagnosis of liver cancer subtypes is challenging.
  • Distinguishing between hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) is critical for treatment.

Purpose of the Study:

  • To develop an automated diagnostic model for liver cancer classification.
  • To differentiate between HCC, ICC, and normal liver tissue using multi-phase CECT images.
  • To create a robust model capable of handling incomplete CECT data.

Main Methods:

  • A Hierarchical Long Short-Term Memory (H-LSTM) model was designed.
  • The model utilizes a shared feature extractor, internal LSTMs for each phase, and an external LSTM across phases.
Keywords:
Contrast-enhanced CT scansDeep learningDiagnostic modelFeature integrationLiver tumor

Related Experiment Videos

Last Updated: Jul 18, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

  • Phase augmentation techniques were applied to enhance model robustness with multi-phase CECT data.

    • Main Results:

    • The H-LSTM model achieved an overall AUROC of 0.93 on the test dataset.
    • Specific AUROC values were 0.97 for HCC and 0.90 for ICC.
    • The model maintained high performance (AUROC > 0.9) even with incomplete CECT phases.

    Conclusions:

    • The H-LSTM model demonstrates high performance for classifying liver cancer subtypes from CECT images, including incomplete datasets.
    • AI-assisted systems show significant potential for improving liver cancer diagnosis and treatment.
    • The H-LSTM model offers a practical and effective solution for clinical diagnostics involving multi-phase imaging data.