Meeting Banner
Abstract #1614

Object Recognition for Fully Automated Reference Tissue Normalization of T2-weighted MR Images of the Prostate

Mattijs Elschot1,2, Gabriel A Nketiah1, Mohammed RS Sunoqrot1, and Tone F Bathen1,2

1Department of Circulation and Medical Imaging, NTNU, Norwegian University of Science and Technology, Trondheim, Norway, 2Department of Radiology and Nuclear Medicine, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway

T2-weighted MRI, an integrated part of multi-parametric MRI for prostate cancer diagnostics, is indispensable for qualitative evaluation of prostate anomalies. For quantitative assessment, however, normalization is necessary for comparison within and between patients. In this study, we developed and validated a fully automated object recognition method for multi-reference tissue normalization. The performance of the method was superior to existing fully automated normalization strategies, and the resulting pseudo T2 values were close to true T2 values from literature. The developed multi-reference tissue normalization method may thus improve the reproducibility and diagnostic performance of T2-weighted image features in future quantitative applications.

This abstract and the presentation materials are available to members only; a login is required.

Join Here