Label fusion Category

4D Multi-atlas Label Fusion using Longitudinal Images

Aug. 29, 2017—Yuankai Huo, Susan M. Resnick and Bennett A. Landman. “4D Multi-atlas Label Fusion using Longitudinal Images”. MICCAI Patch-MI Workshop, 2017. Full text: https://drive.google.com/open?id=0Bzzeqiij2Zara1ZlQXJiclM2UEE Abstract Longitudinal reproducibility is an essential concern in automated medical image segmentation, yet has proven to be an elusive objective as manual brain structure tracings have shown more than 10% variability. To improve reproducibility, longitudinal...

This figure shows the failures for seven problematic volumes. Both whole brain segmentations and cortical surfaces on axial slices are provided. The areas in red rectangles show the global failures while the areas in blue rectangles show the local failures. MaCRUISE did not exhibit such failures in any images.

Combining Multi-atlas Segmentation with Brain Surface Estimation

Mar. 1, 2016—Yuankai Huo, Aaron Carass, Susan M. Resnick, Dzung L. Pham, Jerry L. Prince, Bennett A. Landman.2016 “Combining Multi-atlas Segmentation with Brain Surface Estimation,” In Proceedings of the SPIE Medical Imaging Conference 2016. Oral presentation. Full Text: http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=2506157 Pubmed Abstract Whole brain segmentation (with comprehensive cortical and subcortical labels) and cortical surface reconstruction are two essential...

The three major tasks presented in this paper and the gamut of their observations. The pig cardiac data can be seen in A), the simulated cylinder data can be seen in B), and the sagittal, axial, and coronal cerebellum data can be seen in C). The range of observations is broken up into three classifications. The good classification represents observations that are high quality observations given the original image slice. The bad classification represents observations where the rules were followed but the labeled images are not necessarily close to the ground truth. The ugly classification represents blatant rule breaking and observations that are completely inconsistent with the expected ground truth.

Foibles, Follies, and Fusion: Assessment of Statistical Label Fusion Techniques for Web-Based Collaborations using Minimal Training

Feb. 1, 2011—Andrew J. Asman, Andrew G. Scoggins, Jerry L. Prince, Bennett A. Landman. “Foibles, Follies, and Fusion: Assessment of Statistical Label Fusion Techniques for Web-Based Collaborations using Minimal Training”, In Proceedings of the SPIE Medical Imaging Conference. Lake Buena Vista, Florida, February 2011 PMC3083117† Full text: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3083117/ Abstract Labeling or parcellation of structures of interest on...

Simulations with traditional random raters. Coronal sections of the three-dimensional volume show the high resolution MRI image (A), manually drawn truth model (B), an example delineation from one random traditional rater (C), and the results of a STAPLE recombination of three label sets (D). STAPLER enables fusion of label sets when raters provide only partial datasets, but performance suffers with decreasing overlap (E). With training data (F), STAPLER improved the performance even with each rater labeling only a small portion of the dataset. Box plots in E and F show mean, quartiles, range up to 1.5σ, and outliers. The highlighted plot in E indicates the simulation for which STAPLER was equivalent to STAPLE--i.e., all raters provide a complete set of labels.

Simultaneous Truth and Performance Level Estimation with Incomplete, Over-complete, and Ancillary Data

Mar. 1, 2010—A. Landman, H. Wan, J. Bogovic, and J. L. Prince. “Simultaneous Truth and Performance Level Estimation with Incomplete, Over-complete, and Ancillary Data”, In Proceedings of the SPIE Medical Imaging Conference. San Diego, CA, February 2010 (Oral Presentation) PMC2917119 Full Text:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2917119/ Abstract Image labeling and parcellation are critical tasks for the assessment of volumetric and morphometric...