接触   Contact

Dr. Yago Diez Donoso
Associate Professor, Faculty of Science, Yamagata University, Japan

Medical Image Registration
Yago Diez, VICOROB Research Group, University of Girona, Girona, Spain.
Tohoku University, invited Talk, Sendai, Japan. March 2015

A brief presentation on general concepts of medical image registration including examples from Brain MRI, mammography, breast ABVS and breast MRI.

See presentation

An Experimental Benchmark for Point Set Coarse Matching
Ferran Roure, Yago Diez, Xavi Lladó, Josep Forest, Tomislav Pribanic and Joaquim Salvi
VICOROB Research Group, University of Girona.
Faculty of Electrical Engineering and Computing, University of Zagreb
Visapp 2015 Berlin Germany

In this work we present a new publicly available database that intends to provide valuable tools for researchers working in the coarse matching field.

See presentation

Comparison of Methods for Current-to-Prior Registration of Breast DCE-MRI
Y. Diez1, A. Gubern-Merida1;2, L. Wang3, S. Diekmann3, J. Martí1, B. Platel2, J. Kramme3, and R. Martí1
1: VICOROB group, University of Girona, Catalonia.
2: Dept. Radiology, Radboud Univ. Medical Center, Nijmegen, The Netherlands
3: Fraunhofer MEVIS - Institute for Medical Image Computing, Bremen.
IWDM 14 (Gifu Japan). 2014

Registration is a fundamental part of DCE-MRI image registration as it provides automatic correspondences between current and prior studies. The deformable nature of the breast and differences in acquisition protocols make this a particularly challenging problem.

In this paper we study three registration algorithms (Affine, SyN and Demons) applied to images obtained from clinical practice. Results, shown to be statistically significant indicate how SyN obtains the best results in terms of the two measures considered (landmark based and image metric).

See presentation

An Efficient Method for Surface Registration
Tomislav Pribanic1, Yago Diez2, Sergio Fernandez2 and Joaquim Salvi2.
1Faculty of Electrical Engineering and Computing, University of Zagreb, Unska 3, 10000 Zagreb, Croatia
2Institute of Informatics and Applications, University of Girona, 17071 Girona, Spain.
Visapp 2013 Barcelona (Spain)

3D object data acquired from different viewpoints are usually expressed in different spatial coordinate systems. However, no method assumed the possibility that part of Euclidean parameters could be reliably known in advance.

We argue that it become relatively convenient to include in 3D reconstruction system some inertial sensor which readily provides info about data orientation. We demonstrate a simple, but yet time efficient and accurate registration method.

See presentation
Download pdf

Registration of serial brain MRI scans from multiple sclerosis patients. Analysis of 3D intensity-based methods
Y. Díez, X. Lladó, A. Oliver, R. Martí, E. Roura, M. Cabezas, O. Ganiler, J. Freixenet, (1) J.C. Vilanova, (2), L. Valls (3), Ll. Ramió-Torrentà, (4), D. Pareto, and A. Rovira (5).
(1) Department of Computer Architecture and Technology, University of Girona, Girona, Spain.
(2) Girona Magnetic Resonance Center, Girona, Spain.
(3) Department of Radiology, Dr. Josep Trueta University Hospital, Girona, Spain.
(4) Multiple Sclerosis and Neuroimmunology Unit, Dr. Josep Trueta University Hospital, IdIBGI. Girona, Spain
(5) Magnetic Resonance Unit (Department of Radiology), Vall d'Hebron University Hospital. Barcelona, Spain
Lyon, France. October 2012.

Registration is a crucial task in serial MRI. However, a comprehensive comparison of registration methods does not yet exist. Although many registration methods exist, the most simple ones (Rigid and Affine) are commonly used. A study on the performance of intensity-based registration methods is presented. Insight is provided in different aspects on the quality of the registration methods studied and the improvements brought by non-rigid methods. The potential for state-of-the-art registration methods to help improve semi-automatic and automatic monitoring of MS lesion evolution is established.

See presentation

Comparison of registration methods using mamographic images
Y. Diez, A. Oliver, X. Lladó and R.Martí, University of Girona, Girona, Spain. 2010.
 Icip 2010 Hong Kong, China, 26-29 sept.

This work presents a quantitative evaluation of eight state-of-the art image registration methods applied to mammographic images. These methods range from a global and rigid transformation to local deformable paradigms using various metrics and multi-resolution approaches.

See presentation

Noisy Road Network Matching
Yago Diez (University of Girona, Spain), Mario Lopez (University of Denver, USA) and J. Antoni Sellarès (University of Girona, Spain).
GIScience 2008: Park City, UT, USA

Problem formalization:
- We describe maps using road crossings. Adjacency degrees act as color cathegories.
- Given two sets of road points A and B, |A| < |B|, find all the subsets B' of B that can be expressed as rigid motions* of A.

We want:
- The points to approximately match (fuzzy nature of real data).
- The adjacency degrees to coincide.
- One-to-one matching!

(*) Rigid motion: composition of a translation and a rotation.

See presentation
Download pdf

Colored Point Set Matching under rigid motion in 3D
Yago Diez and J. Antoni Sellarès
University of Girona. Spain. June 25-27, 2007

Problem formulation:
Given two sets of colored points in space A and B, |A| < |B|, find all the subsets B' of B that can be expressed as rigid motions* of A.

We want:
- the points to approximately match (fuzzy nature of real data).
- the colors to coincide.

(*) Rigid motion: composition of a translation and a rotation.

See presentation
Download pdf

2D and 3D Noisy Colored Point Set Matching under the Bottleneck Distance
PhD Thesis presentation, by Yago Diez
Advisor: Dr. Joan Antoni Sellarès
Universitat Politècnica de Catalunya- BarcelonaTech / University of Girona. Spain. October 9th, 2008

Determining the presence of a geometric pattern in a large set of objects is a fundamental problem in computational geometry. We focus in the particular case of Noisy Bottleneck Matching under rigid motion for colored Point sets. We propose several algorithms for determining matches between two colored point sets in two and three dimensions.

Practical aspects are presented from the point of view of some of the implementation details of the algorithms introduced as well as discussion on the experimentation methodology used and its importance in the realization of this thesis.

Download pdf