News Category

Comparison of existing and proposed labeling approaches. In a traditional context, either an expert rater with extensive anatomical knowledge evaluates each dataset (A) or a small set of well-training domain experts who have been instructed by an anatomical expert (B) label each image. Intra- and inter-rater reproducibility analyses are typically performed a per-protocol basis rather than on all datasets. In the proposed WebMILL approach (C), a computer system divides the set of images to be labeled into simple puzzles consisting of a piece of a larger volume and distributes these challenges to a distributed collection of minimally training individuals. Each piece is labeled multiple times by multiple raters. A statistical fusion process simultaneously estimates the true label for each pixel and performance characteristics of each rater.

Foibles, Follies, and Fusion: Web-Based Collaboration for Medical Image Labeling

Jan. 31, 2012—Bennett A Landman, Andrew J Asman, Andrew G Scoggins, John A Bogovic, Joshua A Stein; Jerry L Prince, “Foibles, Follies, and Fusion: Web-Based Collaboration for Medical Image Labeling”, NeuroImage. 2012 Jan 2;59(1):530-9. PMC3195954 † Full text: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3195954/ Abstract Labels that identify specific anatomical and functional structures within medical images are essential to the characterization of...

The inaccuracies of the STAPLE model of rater behavior. A representative slice from the truth model is shown in (A). The expected STAPLE model of rater behavior can be seen in (B). STAPLE operates under the assumption that there is a uniform probability that any given rater would mis-label a given voxel. The observed model of rater behavior can be seen in (C). The primary difference between (B) and (C) is that the human raters showed a clear inclination to mislabel boundary pixels and other ambiguous regions.

Robust Statistical Label Fusion through Consensus Level, Labeler Accuracy and Truth Estimation (COLLATE)

Oct. 31, 2011—Andrew J. Asman and Bennett A. Landman. “Robust Statistical Label Fusion through Consensus Level, Labeler Accuracy and Truth Estimation (COLLATE)”, IEEE Transactions on Medical Imaging. 2011 October; 30(10): 1779–1794. PMC3150602 † Full text: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3150602/ Abstract Segmentation and delineation of structures of interest in medical images is paramount to quantifying and characterizing structural, morphological, and functional...

mpact of outliers in BPM. (A) is one of the regressor images without outliers. The main modality images are: Y = 3X inside a spherical region, Y = constant outside the region. (B) is the Beta estimation when there are no outliers. (C) is the significant results when there are no outliers, (D) is the shifted regressor image, the center cube shifts to another region. The center cube of the paired main modality image shifts to the same region. (E) is the estimated Beta from original BPM when there are outliers. (F) is the significant results from original BPM when there are outliers.

“Robust Biological Parametric Mapping: An Improved Technique for Multimodal Brain Image Analysis”

Feb. 1, 2011—Xue Yang, Lori Beason-Held, Susan M. Resnick, Bennett A. Landman. “Robust Biological Parametric Mapping: An Improved Technique for Multimodal Brain Image Analysis”, In Proceedings of the SPIE Medical Imaging Conference. Lake Buena Vista, Florida, February 2011 PMC3103184 † Full text Abstract Mapping the quantitative relationship between structure and function in the human brain is an...

Interpolation of the genu of the Corpus Callosum (CC) and longitudinal fasciculus
(LF) near the lateral ventricle (V): region of interest (ROI) selected on an
orientation color coded anisotropy map (A,B). MD (C) and log-determinant (D) are
shown at the reconstructed resolution. Upsampled (4 × 4) and interpolated contrasts
are shown for linear (E-G), Log-Euclidean (H-J), and CLEAR (K-M) tensor interpolation.
While the linear interpolation has smooth FA and MD (E,F), it exhibits abrupt
changes in the determinant (G). The Log-Euclidean interpolation exhibits FA artifacts
in regions of high anisotropy and curvature (H). CLEAR avoids all these problems

Tensor Interpolation by Concise Local Estimation of Anisotropy and Rotation

Sep. 1, 2008—A. Landman, P-L. Bazin, and J. L. Prince. “Tensor Interpolation by Concise Local Estimation of Anisotropy and Rotation”, In Proceedings of the Workshop on Computational Diffusion MRI at the 11th International Conference on Medical Image Computing and Computer Assisted Intervention, New York, NY, September 2008. Full Text:http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.701.4262&rep=rep1&type=pdf#page=265 Abstract In diffusion tensor imaging, tensors represent the...

CFARI identified consistent intra-voxel structure in the crossing fibers between
the corpus callosum (CC, lower left, red) and internal capsule (IC, blue, lower right).
Conventional tensor modeling does not preserve the connectivity of the corpus callosum
to the lateral hemispheres

Compressed Sensing of Multiple Intra-Voxel Orientations with Traditional DTI

Sep. 1, 2008—A. Landman, J. Bogovic, and J. L. Prince. “Compressed Sensing of Multiple Intra-Voxel Orientations with Traditional DTI”, In Proceedings of the Workshop on Computational Diffusion MRI at the 11th International Conference on Medical Image Computing and Computer Assisted Intervention, New York, NY, September 2008. Full Text:http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.701.4262&rep=rep1&type=pdf#page=183 Abstract Diffusion tensor imaging (DTI) is widely used to...