Vascular Plaque Determination for Stroke Risk Assessment

Abstract

Diabetic patients are at higher risk of atherosclerotic driven carotid stenosis and subsequent stroke. Point-of-care information regarding plaque composition is needed for all patients, especially diabetic patients. The current research project is a pilot clinical study enrolled subjects who are scheduled for carotid endarterectomy (CEA). A research ultrasound exam of the carotid artery was performed prior to surgery. Following surgery,serial histology was applied to the excised plaque. This histology was used as the gold standard when training the statistical classifiers for estimating plaque composition. During this research effort the key accomplishments are (1) reaching 100% enrollment; (2)demonstrating that spectral information extracted from the nonlinearly generated second harmonic is beneficial for plaque characterization; (3) demonstrating that the reference phantom approach is comparable to other more computationally intensive approaches. The best performing classifier had accuracies of 84%, 74%, and 81% for F, HNC, and Ca respectively.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2021
Accession Number
AD1213957

Entities

People

  • David G. Vince
  • Russell J. Fedewa

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Arteries
  • Cerebrovascular Disorders
  • Data Acquisition
  • Data Mining
  • Data Processing
  • Diffraction
  • Frequency Bands
  • Frequency Shift
  • Health Services
  • Image Processing
  • Information Science
  • Machine Learning
  • Medical Personnel
  • Myocardial Ischemia
  • Power Spectra
  • Risk
  • Scattering
  • Signal Processing
  • Supervised Machine Learning
  • Vascular Diseases

Fields of Study

  • Medicine

Readers

  • Cardiovascular Physiology
  • Medical Imaging.

Technology Areas

  • AI & ML