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 Videos

A multi-resolution approach for spinal metastasis detection using deep Siamese neural networks.

Juan Wang1, Zhiyuan Fang2, Ning Lang3

  • 1Institute for Genomics and Bioinformatics and Department of Computer Science, University of California, Irvine, CA 92697, USA.

Computers in Biology and Medicine
|April 2, 2017
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

Prediction of pedicle screw fixation strength under craniocaudal cyclic load: comparison of various models trained on quantitative CT based finite element analysis.

European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society·2026
Same author

Peripheral trigeminal nerve branch lesions: CT and MRI findings in a pathologically confirmed case series.

BMC surgery·2026
Same author

Spatiotemporal dynamics of the human cortical functional hierarchy across the lifespan.

Nature communications·2026
Same author

Automated diagnostic of cervical spondylosis on multimodal medical images with a multi-task deep learning model.

Nature communications·2026
Same author

Evaluating long-term outcomes of endovascular revascularization in symptomatic chronic internal carotid artery occlusion: A multicenter cohort study.

Clinical neurology and neurosurgery·2026
Same author

Hierarchical maturation of structural brain connectomes from birth to childhood.

Nature communications·2026

This study introduces a Siamese deep neural network for automated spinal metastasis detection in MRI. The method accurately identifies spinal tumors while significantly reducing false positives, improving diagnostic efficiency.

Area of Science:

  • Medical imaging analysis
  • Artificial intelligence in oncology
  • Spinal oncology

Background:

  • Spinal metastasis is the most common spinal malignancy.
  • Accurate detection of spinal metastases in MRI is crucial for patient management.
  • Existing detection methods may face challenges with lesion size variability and false positives.

Purpose of the Study:

  • To investigate the feasibility of automated spinal metastasis detection using deep learning.
  • To develop a robust deep learning model capable of handling diverse metastatic lesion sizes.
  • To reduce false positives in automated spinal metastasis detection.

Main Methods:

  • Development of a Siamese deep neural network with three subnetworks for multi-resolution analysis.
  • Inputting image patches at three different resolutions into the network.
Keywords:
Deep learningMagnetic resonance imagingMulti-resolution analysisSiamese neural networkSpinal metastasis

Related Experiment Videos

  • Utilizing a weighted averaging strategy to aggregate results from neighboring MRI slices for false positive reduction.

    • Main Results:

    • The proposed Siamese neural network correctly detected all spinal metastatic lesions in a set of 26 cases.
    • The method achieved a low false positive rate of 0.40 per case.
    • The aggregation strategy reduced false positives by 44.8% at a 90% true positive rate.

    Conclusions:

    • The Siamese neural network combined with slice aggregation offers a viable strategy for automated spinal metastasis detection in MRI.
    • This approach demonstrates high accuracy and improved specificity.
    • The method has the potential to enhance clinical diagnostic workflows for spinal oncology.