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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Inferring and evaluating network medicine-based disease modules with nextflow.

Bioinformatics (Oxford, England)·2026
Same author

Generating Alzheimer's narratives using large language models.

BMC medical informatics and decision making·2026
Same author

Pushing the boundaries of robotic computed tomography: automated twin-robot CT scan with maximum reachability.

Scientific reports·2026
Same author

Synthesizing vocal tract magnetic resonance imaging sequences with phoneme-aware diffusion models.

Journal of medical imaging (Bellingham, Wash.)·2026
Same author

Illuminating the black box of reservoir computing.

Scientific reports·2026
Same author

Drugst.One DREAM-Drug repurposing through expert annotation and modification.

British journal of pharmacology·2026

Related Experiment Video

Updated: Sep 4, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K

Deep learning for brain metastasis detection and segmentation in longitudinal MRI data.

Yixing Huang1,2, Christoph Bert1,2, Philipp Sommer1,2

  • 1Department of Radiation Oncology, Universitätsklinikum Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, Germany.

Medical Physics
|July 14, 2022
PubMed
Summary
This summary is machine-generated.

This study introduces a novel deep learning approach to improve the detection of brain metastases (BM) in cancer patients. The new method significantly enhances sensitivity and precision, aiding clinicians in diagnosis and treatment planning.

Keywords:
MRIbrain metastasisdeep learningensembleloss functionsensitivity specificity

More Related Videos

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.4K
Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging
06:44

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Published on: June 7, 2020

7.5K

Related Experiment Videos

Last Updated: Sep 4, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.4K
Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging
06:44

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Published on: June 7, 2020

7.5K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence in Medicine
  • Oncology

Background:

  • Brain metastases (BM) are common in metastatic cancer, necessitating early and accurate detection for effective treatment planning and prognosis.
  • Manual detection of small, low-contrast BM is challenging, limiting diagnostic accuracy.
  • Deep learning shows promise for automated BM detection, but sensitivity for tiny lesions and differentiating true positives from false positives (FPs) remain challenges.

Purpose of the Study:

  • To enhance the sensitivity and precision of automated brain metastasis detection using deep learning.
  • To address the challenge of differentiating true metastases from false positives in clinical practice.
  • To improve the integration of automated detection tools into diagnostic and therapeutic workflows for neuroradiologists and radiation oncologists.

Main Methods:

  • A baseline DeepMedic network with binary cross-entropy (BCE) loss was established.
  • A custom volume-level sensitivity-specificity (VSS) loss was developed to optimize detection at a (sub)volume level.
  • A temporal prior volume was incorporated as an additional input to the DeepMedic network (termed DeepMedic+) to reduce false positives.

Main Results:

  • The VSS loss significantly improved sensitivity from 85.3% (BCE) to 97.5% and precision from 69.1% to 98.7%.
  • DeepMedic+ reduced FP metastases by 44.4% in the high-sensitivity model and achieved 99.6% precision in the high-specificity model.
  • An ensemble of high-sensitivity and high-specificity models resulted in an average of only 1.5 FP metastases per patient requiring expert review, with a mean Dice coefficient of 0.81.

Conclusions:

  • The proposed VSS loss and temporal prior effectively enhance brain metastasis detection sensitivity and precision.
  • Ensemble learning successfully distinguishes high-confidence true positives from candidates needing expert review, aligning with clinical needs.
  • This approach facilitates metastasis detection and segmentation, supporting neuroradiologists and radiation oncologists in clinical applications.