Image metric-based correction (autocorrection) of motion effects: Analysisof image metrics

Magnetic resonance (MR) imaging of the shoulder necessitates high spatial a nd contrast resolution resulting in long acquisition times, predisposing th ese images to degradation due to motion. Autocorrection is a new motion cor rection algorithm that attempts to deduce motion during imaging by calculat ing a metric that reflects image quality and searching for motion values th at optimize this metric. The purpose of this work is to report on the evalu ation of 24 metrics for use in autocorrection of MR images of the rotator c uff. Raw data from 164 clinical coronal rotator cuff exams acquired with in terleaved navigator echoes were used. Four observers then scored the origin al and corrected images based on the presence of any motion-induced artifac ts. Changes in metric values before and after navigator-based adaptive moti on correction were correlated with changes in observer score using a least- squares linear regression model. Based on this analysis, the metric that ex hibited the strongest relationship with observer ratings of MR shoulder ima ges was the entropy of the one-dimensional gradient along the phase-encodin g direction, We speculate (and show preliminary evidence) that this metric will be useful not only for autocorrection of shoulder MR images but also f or autocorrection of other MR exams. (C) 2000 Wiley-Liss, Inc.