Automated eye blink detection and correction method for clinical MR eye imaging.

ABSTRACT

PURPOSE: To implement an on-line monitoring system to detect eye blinks during ocular MRI using field probes, and to reacquire corrupted k-space lines by means of an automatic feedback system integrated with the MR scanner. METHODS: Six healthy subjects were scanned on a 7 Tesla MRI whole-body system using a custom-built receive coil. Subjects were asked to blink multiple times during the MR-scan. The local magnetic field changes were detected with an external fluorine-based field probe which was positioned close to the eye. The eye blink produces a field shift greater than a threshold level, this was communicated in real-time to the MR system which immediately reacquired the motion-corrupted k-space lines. RESULTS: The uncorrected images, using the original motion-corrupted data, showed severe artifacts, whereas the corrected images, using the reacquired data, provided an image quality similar to images acquired without blinks. CONCLUSION: Field probes can successfully detect eye blinks during MRI scans. By automatically reacquiring the eye blink-corrupted data, high quality MR-images of the eye can be acquired. Magn Reson Med 78165-171, 2017. © 2016 International Society for Magnetic Resonance in Medicine.