Autocorrection of MR images using multiple metrics

US6265874A · kind A · utility

Assignee

• Mayo Foundation for Medical Education and Research · US

Inventors

• Kiaran P. McGee · Rochester, US
• Joel P. Felmlee · Rochester, US
• Richard L. Ehman · Rochester, US
• Armando Manduca · Rochester, US

Key dates

• Filing date: May 23, 2000
• Grant date: Jul 24, 2001
• Expiry date: May 23, 2020

Classification

• Technology area (CPC G): Physics
• CPC primary: G01R33/5673
• WIPO field: Measurement
• WIPO sector: Instruments

Abstract

An MRI image is corrected for motion artifacts using an iterative, autocorrection process in which corrections are tried and the quality of the resulting reconstructed image is measured. Different metrics for evaluating image quality are used during the autocorrection process to take advantage of their different attributes.