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

Striatal functional connectivity alterations in mild cognitive impairment subtypes defined by CSF A/T biomarkers.

Frontiers in aging neuroscience·2026
Same author

Reversible coma and elevated central venous pressure: warning signs of acute superior vena cava graft thrombosis following Warden procedure: a case report.

BMC anesthesiology·2026
Same author

Preoperative imaging evaluation of primary trigeminal neuralgia using 3D TOF-MRA and 3D FIESTA-c: A retrospective study of 412 cases.

Journal of neurology·2026
Same author

Amygdala subregional atrophy across ATN-defined Mild Cognitive Impairment subgroups.

Frontiers in neuroscience·2026
Same author

Aneurysm Wall Enhancement in Patients With Unruptured Intracranial Aneurysm Using Aspirin or Statins: A Cross-Sectional Analysis of Three Prospective Cohorts.

AJNR. American journal of neuroradiology·2026
Same author

Dosimetric Comparison of Automated Noncoplanar VMAT (HyperArc) Versus CyberKnife for Single-Fraction Vestibular Schwannoma Stereotactic Radiosurgery.

Cancers·2026

Related Experiment Video

Updated: Sep 23, 2025

Digital Spatial Profiling for Characterization of the Microenvironment in Adult-Type Diffusely Infiltrating Glioma
09:17

Digital Spatial Profiling for Characterization of the Microenvironment in Adult-Type Diffusely Infiltrating Glioma

Published on: September 13, 2022

2.4K

Machine Learning Based on Diffusion Kurtosis Imaging Histogram Parameters for Glioma Grading.

Liang Jiang1, Leilei Zhou1, Zhongping Ai1

  • 1Department of Radiology, Nanjing First Hospital, Nanjing Medical University, Nanjing 210029, China.

Journal of Clinical Medicine
|May 14, 2022
PubMed
Summary

Diffusion kurtosis imaging (DKI) histogram parameters, analyzed with support vector machine (SVM) and LASSO, effectively distinguish glioma grades, improving diagnostic accuracy for neuro-oncology.

Keywords:
diffusion kurtosis imaginghistogram analysismachine learningpathological grade

More Related Videos

Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections
05:45

Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections

Published on: July 31, 2017

9.7K
Live Imaging of Microtubule Dynamics in Glioblastoma Cells Invading the Zebrafish Brain
09:29

Live Imaging of Microtubule Dynamics in Glioblastoma Cells Invading the Zebrafish Brain

Published on: July 29, 2022

2.8K

Related Experiment Videos

Last Updated: Sep 23, 2025

Digital Spatial Profiling for Characterization of the Microenvironment in Adult-Type Diffusely Infiltrating Glioma
09:17

Digital Spatial Profiling for Characterization of the Microenvironment in Adult-Type Diffusely Infiltrating Glioma

Published on: September 13, 2022

2.4K
Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections
05:45

Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections

Published on: July 31, 2017

9.7K
Live Imaging of Microtubule Dynamics in Glioblastoma Cells Invading the Zebrafish Brain
09:29

Live Imaging of Microtubule Dynamics in Glioblastoma Cells Invading the Zebrafish Brain

Published on: July 29, 2022

2.8K

Area of Science:

  • Radiology
  • Neuro-oncology
  • Medical Imaging Analysis

Background:

  • Glioma grading is crucial for surgical planning and patient prognosis.
  • Accurate differentiation between low-grade and high-grade gliomas remains a clinical challenge.
  • Diffusion kurtosis imaging (DKI) offers advanced insights into tissue microstructure beyond conventional MRI.

Purpose of the Study:

  • To evaluate the efficacy of feature reduction methods combined with support vector machine (SVM) classifiers for glioma grading using DKI histogram parameters.
  • To compare the performance of different feature selection techniques (PCA, RFE, LASSO) in differentiating glioma grades.
  • To assess the superiority of DKI-based machine learning models over conventional statistical methods for glioma grading.

Main Methods:

  • Retrospective analysis of 161 glioma patients' MRI data, divided into low-grade (n=61) and high-grade (n=100) groups.
  • Derivation of parametric DKI maps and semi-automatic extraction of 45 histogram features.
  • Application of three feature selection methods (PCA, RFE, LASSO) to build SVM-based glioma grading models.

Main Results:

  • Conventional ROC analysis identified mean diffusivity (MD) variance, MD skewness, and mean kurtosis (MK) C50 as key discriminators, with MD variance being particularly effective.
  • The highest classification accuracy (AUC = 0.904 ± 0.069) was achieved using the LASSO feature selection method with SVM.
  • The SVM-PCA model, despite having the lowest AUC among SVM models, demonstrated significantly better performance than conventional ROC analysis (p=0.013).

Conclusions:

  • DKI histogram parameters, when analyzed using LASSO and SVM, provide a robust and superior method for distinguishing glioma grades.
  • Machine learning approaches applied to DKI data enhance diagnostic accuracy in neuro-oncology.
  • This study highlights the potential of advanced MRI techniques and computational analysis for improved glioma grading and patient management.