Meeting Banner
Abstract #1254

Histology Assisted Validation of Automatic Detection of Soft Plaque in Vessel Wall Images by Using Optimal Number of MR Sequences

Ronald van 't Klooster1, Andrew J. Patterson2, Victoria E. Young2, Jonathan H. Gillard2, Johan H.C. Reiber1, Rob J. van der Geest1

1Division of Image Processing, Department of Radiology, Leiden University Medical Center, Leiden, Netherlands; 2University Department of Radiology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, United Kingdom

Extensive MR vessel wall imaging protocols are used to identify unstable plaques, which play an important role in the progression of atherosclerosis. Comparison was made between automatic plaque detection, by a supervised classification system, and histology assisted manual segmentation. Experiments show that the automatic detection of unstable plaque is in good agreement with the manual segmentation. Moreover, the STIR and DWI sequences show an improvement over the T2w and PDw sequences. Automatic detection of soft plaque may be feasible by using a limited number of MR sequences, saving both MRI system and image analysis time.