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 15, 2026

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

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

Published on: September 25, 2019

Efficient brain lesion segmentation using multi-modality tissue-based feature selection and support vector machines.

Jean-Baptiste Fiot¹, Laurent D Cohen, Parnesh Raniga

¹CEREMADE, UMR 7534 CNRS Université Paris Dauphine, France; CSIRO Preventative Health National Research Flagship ICTC, The Australian e-Health Research Centre - BioMedIA, Royal Brisbane and Women's Hospital, Herston, Qld, Australia.

International Journal for Numerical Methods in Biomedical Engineering

|January 11, 2013

Summary

This summary is machine-generated.

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

Predicting accumulation and age at onset of amyloid-β from genetic risk and resilience for Alzheimer's disease.

EBioMedicine·2026

Same author

Feasibility and Acceptability of a Remote Sleep-Dependent Memory Assessment in Older Adults With Cognitive Concerns: Pilot Cross-Sectional Study.

JMIR aging·2026

Same author

Comparative Diagnostic Performance of Early and Term MRI in Preterm Infants: a Diagnostic Test Accuracy Systematic review and Bayesian Meta-analysis.

Neonatology·2026

Same author

Multimodal ultra-high-field MRI, clinical, cognitive, and genetic profiles across the ALS-FTD spectrum.

Scientific data·2026

Same author

Interpretable Semantic Medical Image Segmentation with Style and Confidence.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Vision Intervention for Seeing Impaired Babies: Learning through Enrichment (VISIBLE) - protocol of a feasibility pilot randomised controlled trial.

BMJ open·2026

Same journal

Dynamic Simulation of Male Bladder Outlet Obstruction: Flow Characteristics and Novel Quantitative Indicators.

International journal for numerical methods in biomedical engineering·2026

Same journal

Path Planning of Flexible Needle Based on Improved Particle Swarm Optimization Algorithm.

International journal for numerical methods in biomedical engineering·2026

Same journal

A Fully Automated Pipeline for Vertebral Structural Assessment From Medical Images. Application Under Metastatic Conditions.

International journal for numerical methods in biomedical engineering·2026

Same journal

Evaluation of Degradable Mg-Alloy Implants for Femoral Neck Fractures: Subject-Specific Finite Element Analysis Integrated With Bone Remodelling.

International journal for numerical methods in biomedical engineering·2026

Same journal

Patient-Specific In Silico Prediction of Outcomes of Partial Continuous-Flow LVAD Treatment in Peripartum Cardiomyopathy.

International journal for numerical methods in biomedical engineering·2026

Same journal

The Impact of Rotational Speed and Descending Aortic Flow on Hemodynamics of a Modular Intra-Aortic Entrainment Pump: A Numerical Study.

International journal for numerical methods in biomedical engineering·2026

See all related articles

This study introduces an improved method for segmenting white matter hyper-intensities (WMH) using support vector machines (SVM). The new approach enhances accuracy and efficiency by optimizing feature selection and pre-processing, reducing computational load.

Area of Science:

Medical Image Analysis
Machine Learning in Medicine
Neurology

Background:

White matter hyper-intensities (WMH) are common in aging and various neurological diseases.
Accurate segmentation of WMH is crucial for diagnosis and monitoring disease progression.
Current support vector machine (SVM) methods for WMH segmentation often require extensive feature extraction and complex post-processing.

Purpose of the Study:

To develop an efficient and accurate method for segmenting white matter hyper-intensities (WMH) using SVM.
To investigate the impact of different MRI modalities, feature neighborhood sizes, and multi-scale features on WMH segmentation accuracy.
To optimize the SVM classification process by focusing on advanced pre-processing and tissue-based feature selection.

Main Methods:

Keywords:

brain lesion classification image processing segmentation support vector machines

Related Experiment Videos

Last Updated: May 15, 2026

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

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

Published on: September 25, 2019

Employed advanced pre-processing techniques and tissue-based feature selection for SVM classification.
Compared WMH segmentation performance using T1-weighted, T2-weighted, proton-density, and fluid-attenuated inversion recovery (FLAIR) MRI sequences.
Evaluated the influence of different neighborhood sizes (e.g., 5x5x5 vs. 3x3x3) and multi-scale features on classification accuracy.

Main Results:

WMH segmentation using all four MRI modalities achieved the highest average Dice scores across small, moderate, and severe lesion loads.
No significant difference in segmentation performance was observed between using all four modalities versus using only T1-w and FLAIR sequences (p=0.50).
Negligible differences in performance were found between 5x5x5 and 3x3x3 feature neighborhood sizes (p=0.93), and the proposed method outperformed a full-feature approach with post-processing.

Conclusions:

Optimized feature selection and pre-processing significantly improve the efficiency and accuracy of SVM-based WMH segmentation.
T1-w and FLAIR sequences alone provide comparable WMH segmentation results to using all four modalities, simplifying the process.
The developed method offers a computationally efficient and accurate solution for WMH segmentation, reducing storage and processing time.