This is the public Matlab implementation for medical image soft-segmentation using the supervised multi-atlas-based expected label value (ELV) approach proposed by Aganj and Fischl (IEEE ISBI 2019; bioRxiv 2020). This approach considers the probability of all possible atlas-to-image transformations and computes the ELV, thus bypassing deformable registration and avoiding the associated computational costs. A short tutorial is included in EXAMPLE.m.
This package also includes functions for FFT-based convolution, which can be independently used.
Iman Aganj (2020). Expected Label Value (ELV) for Image Segmentation (https://www.mathworks.com/matlabcentral/fileexchange/81283-expected-label-value-elv-for-image-segmentation), MATLAB Central File Exchange. Retrieved .
Aganj, Iman, and Bruce Fischl. Multi-Atlas Image Soft-Segmentation via Computation of the Expected Label Value. Cold Spring Harbor Laboratory, Oct. 2020, doi:10.1101/2020.10.08.331553.
Aganj, Iman, and Bruce Fischl. “Expected Label Value Computation for Atlas-Based Image Segmentation.” 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), IEEE, 2019, doi:10.1109/isbi.2019.8759484.
Find the treasures in MATLAB Central and discover how the community can help you!Start Hunting!