Meeting Banner
Abstract #2859

Generic feature extraction accompanied by support vector classification: an efficient and effective way for MR image quality determination

Dirk Bequé1, Arathi Sreekumari2, Dattesh Shanbhag2, Keith Park3, Desmond Teck Beng Yeo3, Thomas K.F. Foo3, and Ileana Hancu3

1GE Global Research, Garching bei München, Germany, 2GE Global Research, Bangalore, India, 3GE Global Research, Niskayuna, NY, United States

Support vector machine image classification is performed on MR brain images to determine the need to repeat the MR acquisition. However, the image feature extraction is completely brain image agnostic. It is performed either directly on image slices or simple transformations thereof, like e.g. by fore/background thresholding or 1-level wavelet decomposition. 120 image features and meta-data entries are used to classify images as sufficient to diagnose or not. 84% accuracy is demonstrated, even after reducing the feature space to only 20 features. Such feature computation is fast enough to perform image quality assessment in real time, immediately after scan completion.

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

Join Here