Self-Aligning Manifolds for Matching Disparate Medical Image Datasets

Christian F. Baumgartner; Alberto Gomez; Lisa M. Koch; James R. Housden; Christoph Kolbitsch; Jamie R. Mcclelland; Daniel Rueckert; Andy P. King

doi:10.1007/978-3-319-19992-4_28

Self-Aligning Manifolds for Matching Disparate Medical Image Datasets

Christian F. Baumgartner, Alberto Gomez, Lisa M. Koch, James R. Housden, Christoph Kolbitsch, Jamie R. Mcclelland, Daniel Rueckert, Andy P. King

Research output: Chapter in Book/Report/Conference proceeding › Other chapter contribution › peer-review

14 Citations (Scopus)

Abstract

Manifold alignment can be used to reduce the dimensionality of multiple medical image datasets into a single globally consistent low-dimensional space. This may be desirable in a wide variety of problems, from fusion of different imaging modalities for Alzheimer’s disease classification to 4DMR reconstruction from 2D MR slices. Unfortunately, most existing manifold alignment techniques require either a set of prior correspondences or comparability between the datasets in high-dimensional space, which is often not possible. We propose a novel technique for the ‘self-alignment’ of manifolds (SAM) from multiple dissimilar imaging datasets without prior correspondences or inter-dataset image comparisons. We quantitatively evaluate the method on 4DMR reconstruction from realistic, synthetic sagittal 2D MR slices from 6 volunteers and real data from 4 volunteers. Additionally, we demonstrate the technique for the compounding of two free breathing 3D ultrasound views from one volunteer. The proposed method performs significantly better for 4DMR reconstruction than state-of-the-art image-based techniques.

Original language	English
Title of host publication	Information Processing in Medical Imaging
Pages	363-374
Volume	90123
DOIs	https://doi.org/10.1007/978-3-319-19992-4_28
Publication status	Published - 23 Jun 2015

Publication series

Name	Lecture notes in Computer Science

Access to Document

10.1007/978-3-319-19992-4_28

http://link.springer.com/10.1007/978-3-319-19992-4_28

Cite this

Baumgartner, C. F., Gomez, A., Koch, L. M., Housden, J. R., Kolbitsch, C., Mcclelland, J. R., Rueckert, D., & King, A. P. (2015). Self-Aligning Manifolds for Matching Disparate Medical Image Datasets. In Information Processing in Medical Imaging (Vol. 90123, pp. 363-374). Article Chapter 28 (Lecture notes in Computer Science). https://doi.org/10.1007/978-3-319-19992-4_28

@inbook{050efc71704e4309aef89061a8ce3cf4,

title = "Self-Aligning Manifolds for Matching Disparate Medical Image Datasets",

abstract = "Manifold alignment can be used to reduce the dimensionality of multiple medical image datasets into a single globally consistent low-dimensional space. This may be desirable in a wide variety of problems, from fusion of different imaging modalities for Alzheimer{\textquoteright}s disease classification to 4DMR reconstruction from 2D MR slices. Unfortunately, most existing manifold alignment techniques require either a set of prior correspondences or comparability between the datasets in high-dimensional space, which is often not possible. We propose a novel technique for the {\textquoteleft}self-alignment{\textquoteright} of manifolds (SAM) from multiple dissimilar imaging datasets without prior correspondences or inter-dataset image comparisons. We quantitatively evaluate the method on 4DMR reconstruction from realistic, synthetic sagittal 2D MR slices from 6 volunteers and real data from 4 volunteers. Additionally, we demonstrate the technique for the compounding of two free breathing 3D ultrasound views from one volunteer. The proposed method performs significantly better for 4DMR reconstruction than state-of-the-art image-based techniques.",

author = "Baumgartner, {Christian F.} and Alberto Gomez and Koch, {Lisa M.} and Housden, {James R.} and Christoph Kolbitsch and Mcclelland, {Jamie R.} and Daniel Rueckert and King, {Andy P.}",

year = "2015",

month = jun,

day = "23",

doi = "10.1007/978-3-319-19992-4_28",

language = "English",

volume = "90123",

series = "Lecture notes in Computer Science",

pages = "363--374",

booktitle = "Information Processing in Medical Imaging",

}

TY - CHAP

T1 - Self-Aligning Manifolds for Matching Disparate Medical Image Datasets

AU - Baumgartner, Christian F.

AU - Gomez, Alberto

AU - Koch, Lisa M.

AU - Housden, James R.

AU - Kolbitsch, Christoph

AU - Mcclelland, Jamie R.

AU - Rueckert, Daniel

AU - King, Andy P.

PY - 2015/6/23

Y1 - 2015/6/23

N2 - Manifold alignment can be used to reduce the dimensionality of multiple medical image datasets into a single globally consistent low-dimensional space. This may be desirable in a wide variety of problems, from fusion of different imaging modalities for Alzheimer’s disease classification to 4DMR reconstruction from 2D MR slices. Unfortunately, most existing manifold alignment techniques require either a set of prior correspondences or comparability between the datasets in high-dimensional space, which is often not possible. We propose a novel technique for the ‘self-alignment’ of manifolds (SAM) from multiple dissimilar imaging datasets without prior correspondences or inter-dataset image comparisons. We quantitatively evaluate the method on 4DMR reconstruction from realistic, synthetic sagittal 2D MR slices from 6 volunteers and real data from 4 volunteers. Additionally, we demonstrate the technique for the compounding of two free breathing 3D ultrasound views from one volunteer. The proposed method performs significantly better for 4DMR reconstruction than state-of-the-art image-based techniques.

AB - Manifold alignment can be used to reduce the dimensionality of multiple medical image datasets into a single globally consistent low-dimensional space. This may be desirable in a wide variety of problems, from fusion of different imaging modalities for Alzheimer’s disease classification to 4DMR reconstruction from 2D MR slices. Unfortunately, most existing manifold alignment techniques require either a set of prior correspondences or comparability between the datasets in high-dimensional space, which is often not possible. We propose a novel technique for the ‘self-alignment’ of manifolds (SAM) from multiple dissimilar imaging datasets without prior correspondences or inter-dataset image comparisons. We quantitatively evaluate the method on 4DMR reconstruction from realistic, synthetic sagittal 2D MR slices from 6 volunteers and real data from 4 volunteers. Additionally, we demonstrate the technique for the compounding of two free breathing 3D ultrasound views from one volunteer. The proposed method performs significantly better for 4DMR reconstruction than state-of-the-art image-based techniques.

U2 - 10.1007/978-3-319-19992-4_28

DO - 10.1007/978-3-319-19992-4_28

M3 - Other chapter contribution

VL - 90123

T3 - Lecture notes in Computer Science

SP - 363

EP - 374

BT - Information Processing in Medical Imaging

ER -

Self-Aligning Manifolds for Matching Disparate Medical Image Datasets

Abstract

Publication series

Access to Document

Fingerprint

Cite this