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 Video

Updated: May 23, 2026

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
15:48

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Published on: December 15, 2014

Comparative analysis of nonlinear dimensionality reduction techniques for breast MRI segmentation.

Alireza Akhbardeh1, Michael A Jacobs

  • 1The Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA.

Medical Physics
|April 10, 2012
PubMed
Summary

A novel hybrid machine learning approach effectively segments breast MRI data, improving visualization and differentiation of tissue types. This method integrates multiple MRI parameters into a single image for enhanced diagnostic insights.

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

The Association Between the Risk Analysis Index of Frailty and Patient Safety Indicators: A Single-System Cohort Study.

The Journal of surgical research·2026
Same author

Objective dyspnea measurement in real time using a forehead wearable and artificial intelligence.

NPJ primary care respiratory medicine·2026
Same author

Prospective clinical study for assessing real-time objective pain using a signal processing algorithm in conjunction with wearable forehead sensors.

Journal of clinical anesthesia·2026
Same author

MRI-Determined Muscle Fat Fractions and Contractile Volumes in Myotonic Dystrophy Type 2.

Muscle & nerve·2026
Same author

Predicting Protracted Recovery and Long-Term Loss of Independence after Major Surgery in Veterans.

Annals of surgery·2026
Same author

Critical Flaws in a Recent JACS Publication Describing Trajectories of Postoperative Recovery.

Journal of the American College of Surgeons·2026

Area of Science:

  • Medical Imaging and Radiology
  • Machine Learning in Healthcare
  • Biomedical Data Analysis

Background:

  • Radiological imaging is crucial for distinguishing normal from pathological tissues.
  • Integrating large datasets from medical imaging is challenging and time-consuming for radiologists.
  • Advanced machine learning offers potential for improved visualization and segmentation of radiological data.

Purpose of the Study:

  • To apply a novel hybrid scheme combining wavelet transform and nonlinear dimensionality reduction (NLDR) for breast MRI data visualization and segmentation.
  • To compare the performance of NLDR techniques (ISOMAP, LLE, DfM) against linear methods (PCA, MDS).
  • To develop an integrated 'embedded image' from multiple MRI parameters for enhanced tissue differentiation.

Main Methods:

Related Experiment Videos

Last Updated: May 23, 2026

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
15:48

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Published on: December 15, 2014

  • A hybrid scheme involving preprocessing (B(1) inhomogeneity correction, registration, wavelet compression) and postprocessing was developed.
  • Nonlinear dimensionality reduction (NLDR) techniques integrated multiple MRI parameters (T1, T2, DWI, DCE) into a single embedded image.
  • Validation involved synthetic data comparison with PCA/MDS and clinical application on breast MRI data, evaluating lesion segmentation congruence.

Main Results:

  • The NLDR-based hybrid approach successfully segmented both synthetic and clinical breast MRI data.
  • NLDR methods achieved high accuracy in segmenting different breast tissue types, outperforming PCA and MDS.
  • The embedded image revealed fuzzy boundaries between tissue types (fatty, glandular, lesional) with >86% accuracy.

Conclusions:

  • The proposed hybrid NLDR methods accurately segment clinical breast MRI data.
  • This approach constructs an embedded image that effectively visualizes the contribution of various radiological parameters.
  • The technique shows promise for improving the analysis and interpretation of complex medical imaging data.