Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: May 24, 2026

Manual Segmentation of the Human Choroid Plexus Using Brain MRI

Manual Segmentation of the Human Choroid Plexus Using Brain MRI

Published on: December 15, 2023

Physics-informed DynUNet for brain metastasis segmentation.

Murat Güzel¹, Ömer Kaan Baykan²

¹Department of Computer Technology, Mucur Vocational School, Kırşehir Ahi Evran University, Kırşehir, Türkiye.

Computer Methods and Programs in Biomedicine

|May 22, 2026

Summary

Physics-informed deep learning improves brain metastasis segmentation, especially for small lesions. Optimal configurations vary by clinical application, offering context-specific deployment for better neuro-oncology outcomes.

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

Can Pleth Variability Index Levels in Patients with Upper Gastrointestinal Bleeding Predict the Requirement for Transfusion.

Clinical laboratory·2026

Same author

Author Response to Letter Regarding "Acute Spinal Epidural Hematoma: A Serious Complication of Chiropractic Therapy-A Case Report".

Journal of emergency nursing·2026

Same author

Profilin-1 as a Potential Biomarker in Acute Ischemic Stroke: Predictive Value for Large Vessel Occlusion and Mechanical Thrombectomy.

Clinical laboratory·2026

Same author

The Role of VEGF-A in Young Patients with Acute Coronary Syndrome.

Clinical laboratory·2026

Same author

Pleth variability index as a non-invasive prognostic indicator in acute exacerbation of chronic obstructive pulmonary disease.

BMC pulmonary medicine·2026

Same author

Initial Lactate vs. Lactate Clearance in Predicting Hyperbaric Oxygen Therapy Requirement in Carbon Monoxide Poisoning.

Clinical laboratory·2026

Area of Science:

Neuro-oncology
Artificial Intelligence
Medical Imaging

Background:

Detecting and segmenting small brain metastases is challenging for standard deep learning.
Existing models lack biological information on metastasis growth, limiting performance on low-volume lesions.

Purpose of the Study:

To investigate if integrating physics-informed (PI) tumor growth models into segmentation architectures can overcome size-dependent limitations in brain metastasis detection.
To evaluate the effectiveness of PI-DynUNet in segmenting brain metastases of varying sizes and characteristics.

Main Methods:

Developed PI-DynUNet by integrating a physics-based growth model with DynUNet.
Trained and compared PI-DynUNet against U-Net variants on the BraTS-METS 2023 dataset using controlled settings.

Keywords:

Brain metastasis Deep learning MRI segmentation Medical image analysis Physics-informed neural networks Tumor boundary delineation

More Related Videos

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Published on: June 7, 2020

Related Experiment Videos

Last Updated: May 24, 2026

Manual Segmentation of the Human Choroid Plexus Using Brain MRI

Manual Segmentation of the Human Choroid Plexus Using Brain MRI

Published on: December 15, 2023

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging

Published on: June 7, 2020

Evaluated performance across six lesion-size categories and assessed clinical context-specific performance using scenario-weighted Dice coefficients.

Main Results:

PI-DynUNet demonstrated improved segmentation across all BraTS regions, with notable gains in non-enhancing tumor core (NETC) Dice (+5.3%).
Optimal regularization weights (λ) were found to be tissue and lesion-size dependent.
Context-dependent optimal models were identified, with PI-DynUNet (λ=0.01) excelling in enhancing-weighted scenarios and λ=1.0 showing superior balanced accuracy.

Conclusions:

Physics-informed deep learning offers modest but significant improvements in brain metastasis segmentation.
External validation confirmed the outperformance of PI-DynUNet over baseline DynUNet, with substantial gains in tumor-core Dice.
Optimal PI-DynUNet configuration is application-specific, guiding context-aware deployment for neuro-oncology applications.