Liver disease, especially Non-Alcoholic Fatty Liver Disease has reached high levels, and there is a need for non-invasive tests based on quantitative MRI to replace biopsy in order to better assess liver health. An automated quantitative liver segmentation approach is required to automate these tests and in this work we propose a fully convolutional framework with a novel objective function for quantitative liver segmentation. The method has (to date) been tested on quantitative T1 maps generated from the UK Biobank study. We obtained extremely encouraging results on an unseen test set with a Dice score of 0.95, and Sensitivity 0.98 and Specificity 0.99.
CITATION STYLE
Irving, B., Hutton, C., Dennis, A., Vikal, S., Mavar, M., Kelly, M., & Brady, S. J. M. (2017). Deep quantitative liver segmentation and vessel exclusion to assist in liver assessment. In Communications in Computer and Information Science (Vol. 723, pp. 663–673). Springer Verlag. https://doi.org/10.1007/978-3-319-60964-5_58
Mendeley helps you to discover research relevant for your work.