Multimodal non-rigid registration methods based on local variability measures in computed tomography and magnetic resonance brain images

Isnardo Reducindo, Elisa Scalco, Edgar R. Arce-Santana, Giovanni M. Cattaneo, Aldo R. Mejia-Rodriguez, Daniel U. Campos-Delgado, Anna M. Bianchi, Giovanna Rizzo, Flavio Vigueras-Gomez
  • IET Image Processing, December 2014, the Institution of Engineering and Technology (the IET)
  • DOI: 10.1049/iet-ipr.2013.0705

Multimodal non-rigid registration

What is it about?

Mathematical method to align medical images of different modalities. The proposed method considers a two step procedure which allows independent motion of the pixels. The paper presents the application to track anatomical structures (brain, tumour and cerebral ventricles) in radiotherapy.

Why is it important?

The proposed methodology allows the alignment of medial images of different modalities and with independent motion of the pixels.


Dr Daniel Ulises Campos-Delgado
Universidad Autonoma de San Luis Potosi

Our multimodal alignment methodology is based on transforming the studied images to a domain where their intensities could be compared, and we called this mapping --> local variability measure. Our method was applied to a set of magnetic resonance/computed tomography clinical studies (pre- and post-radiotherapy treatment) of three patients with cerebral tumour deformations of the brain structures. This paper was made possible by collaborative work between institutions from Mexico and Italy.

Read Publication

The following have contributed to this page: Dr Daniel Ulises Campos-Delgado