Raid 4D MRI of GAD-DTPA Enhancement for Breast Lesion Characterization

Abstract

The aims of this project were refinement and implementation of new methods designed to defeat the tradeoff between spatial resolution, temporal resolution and tissue coverage in dynamic gadolinium-enhanced MR mammography. The approach involved specialized software for data acquisition and reconstruction to yield dynamic scans of 32 sections through both breasts acquired at 5-6 second temporal resolution. Additional post-processing reduced motion artifacts for improved quantitation via in numerical fitting of temporal enhancement properties of tissues. Technical objectives related to completion of data acquisition and processing methods were generally successful. This methodology was applied in a prospective 4-year study of 102 women referred for surgical biopsy of a breast abnormality detected by non-MRI means. The hypothesis that malignancies to will display a high contrast-enhancement rate was supported by the data RATE = (79 +/- 75)X10(EXP -3)/SEC, MEAN +/- STD DEV, N=39 relative to benign tissues RATE = (19 +/- 16)X10(EXP -3)/SEC, N=54, although some overlap exists. ROC analysis yields a sensitivity 85% and specificity 75% using an enhancement-rate threshold = 22x10(exp -3)/sec. These quantitative kinetic properties can be further combined with tissue/lesion morphology to further enhance specificity of breast MRI.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 01, 1999
Accession Number
ADA370330

Entities

People

  • Thomas L. Chenevert

Organizations

  • University of Michigan

Tags

DTIC Thesaurus Topics

  • Abnormalities
  • Acquisition
  • Amplitude
  • Breast Cancer
  • Cancer
  • Carcinoma
  • Data Acquisition
  • Data Sets
  • Detection
  • Diseases And Disorders
  • Health Services
  • Magnetic Resonance
  • Magnetic Resonance Imaging
  • Medical Personnel
  • Neoplasms
  • Three Dimensional
  • Two Dimensional

Fields of Study

  • Physics

Readers

  • Mathematics or Statistics
  • Medical Imaging.
  • Oncology and Biomarker-Based Cancer Detection.