Incorporating non-rigid registration into expectation maximization algorithm to segment MR images

K. M. Pohl, W. M. Wells III, A. Guimond, K. Kasai, M. E. Shenton, R. Kikinis, W. E. L. Grimson, S. K. Warfield
Proceedings of the 5th International Conference on Medical Image Computing and Computer Assisted Intervention MICCAI 2002
Pages 564-572
2002

Download full paper

Abstract

The paper introduces an algorithm which allows the automatic segmentation of multi channel magnetic resonance images. We extended the Expectation Maximization-Mean Field Approximation Segmenter, to include Local Prior Probability Maps. Thereby our algorithm estimates the bias field in the image while simultaneously assigning voxels to different tissue classes under prior probability maps. The probability maps were aligned to the subject using non-rigid registration. This allowed the parcellation of cortical sub-structures including the superior temporal gyrus. To our knowledge this is the first description of an algorithm capable of automatic cortical parcellation incorporating strong noise reduction and image intensity correction.

---

Reference

Pohl KM, III WMW, Guimond A, Kasai K, Shenton ME, Kikinis R, Grimson WEL, Warfield SK. Incorporating non-rigid registration into expectation maximization algorithm to segment mr images. In T Dohi, R Kikinis, eds., Proceedings of the 5th International Conference on Medical Image Computing and Computer Assisted Intervention MICCAI 2002. Springer-Verlag GmbH, 2002;564-572.

Grants

NIH P41 RR13218, NIH P01-CA67165, NIH R01 RR11747, NIH RO1 CA86879, Whitaker Foundation, New Concept Award

Research area

shapeanalysis