Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition

Ke Zeng, Guray Erus, Aristeidis Sotiras, Russell T. Shinohara, Christos Davatzikos

Research output: Contribution to journalArticle

Abstract

We present a generic method for automatic detection of abnormal regions in medical images as deviations from a normative data base. The algorithm decomposes an image, or more broadly a function defined on the image grid, into the superposition of a normal part and a residual term. A statistical model is constructed with regional sparse learning to represent normative anatomical variations among a reference population (e.g., healthy controls), in conjunction with a Markov random field regularization that ensures mutual consistency of the regional learning among partially overlapping image blocks. The decomposition is performed in a principled way so that the normal part fits well with the learned normative model, while the residual term absorbs pathological patterns, which may then be detected through a statistical significance test. The decomposition is applied to multiple image features from an individual scan, detecting abnormalities using both intensity and shape information. We form an iterative scheme that interleaves abnormality detection with deformable registration, gradually improving robustness of the spatial normalization and precision of the detection. The algorithm is evaluated with simulated images and clinical data of brain lesions, and is shown to achieve robust deformable registration and localize pathological regions simultaneously. The algorithm is also applied on images from Alzheimer's disease patients to demonstrate the generality of the method.

Original languageEnglish (US)
Article number7426805
Pages (from-to)1937-1951
Number of pages15
JournalIEEE Transactions on Medical Imaging
Volume35
Issue number8
DOIs
StatePublished - Aug 1 2016
Externally publishedYes

Fingerprint

Decomposition
Learning
Statistical tests
Statistical Models
Brain
Alzheimer Disease
Databases
Population

Keywords

  • Abnormality detection
  • basis pursuit
  • brain pathology
  • convex optimization
  • medical image registration
  • sparse representation

ASJC Scopus subject areas

  • Software
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition. / Zeng, Ke; Erus, Guray; Sotiras, Aristeidis; Shinohara, Russell T.; Davatzikos, Christos.

In: IEEE Transactions on Medical Imaging, Vol. 35, No. 8, 7426805, 01.08.2016, p. 1937-1951.

Research output: Contribution to journalArticle

Zeng, K, Erus, G, Sotiras, A, Shinohara, RT & Davatzikos, C 2016, 'Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition', IEEE Transactions on Medical Imaging, vol. 35, no. 8, 7426805, pp. 1937-1951. https://doi.org/10.1109/TMI.2016.2538998
Zeng, Ke ; Erus, Guray ; Sotiras, Aristeidis ; Shinohara, Russell T. ; Davatzikos, Christos. / Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition. In: IEEE Transactions on Medical Imaging. 2016 ; Vol. 35, No. 8. pp. 1937-1951.
@article{03284ce4f3fe44d3a04bb72c25c3d3e8,
title = "Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition",
abstract = "We present a generic method for automatic detection of abnormal regions in medical images as deviations from a normative data base. The algorithm decomposes an image, or more broadly a function defined on the image grid, into the superposition of a normal part and a residual term. A statistical model is constructed with regional sparse learning to represent normative anatomical variations among a reference population (e.g., healthy controls), in conjunction with a Markov random field regularization that ensures mutual consistency of the regional learning among partially overlapping image blocks. The decomposition is performed in a principled way so that the normal part fits well with the learned normative model, while the residual term absorbs pathological patterns, which may then be detected through a statistical significance test. The decomposition is applied to multiple image features from an individual scan, detecting abnormalities using both intensity and shape information. We form an iterative scheme that interleaves abnormality detection with deformable registration, gradually improving robustness of the spatial normalization and precision of the detection. The algorithm is evaluated with simulated images and clinical data of brain lesions, and is shown to achieve robust deformable registration and localize pathological regions simultaneously. The algorithm is also applied on images from Alzheimer's disease patients to demonstrate the generality of the method.",
keywords = "Abnormality detection, basis pursuit, brain pathology, convex optimization, medical image registration, sparse representation",
author = "Ke Zeng and Guray Erus and Aristeidis Sotiras and Shinohara, {Russell T.} and Christos Davatzikos",
year = "2016",
month = "8",
day = "1",
doi = "10.1109/TMI.2016.2538998",
language = "English (US)",
volume = "35",
pages = "1937--1951",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "8",

}

TY - JOUR

T1 - Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition

AU - Zeng, Ke

AU - Erus, Guray

AU - Sotiras, Aristeidis

AU - Shinohara, Russell T.

AU - Davatzikos, Christos

PY - 2016/8/1

Y1 - 2016/8/1

N2 - We present a generic method for automatic detection of abnormal regions in medical images as deviations from a normative data base. The algorithm decomposes an image, or more broadly a function defined on the image grid, into the superposition of a normal part and a residual term. A statistical model is constructed with regional sparse learning to represent normative anatomical variations among a reference population (e.g., healthy controls), in conjunction with a Markov random field regularization that ensures mutual consistency of the regional learning among partially overlapping image blocks. The decomposition is performed in a principled way so that the normal part fits well with the learned normative model, while the residual term absorbs pathological patterns, which may then be detected through a statistical significance test. The decomposition is applied to multiple image features from an individual scan, detecting abnormalities using both intensity and shape information. We form an iterative scheme that interleaves abnormality detection with deformable registration, gradually improving robustness of the spatial normalization and precision of the detection. The algorithm is evaluated with simulated images and clinical data of brain lesions, and is shown to achieve robust deformable registration and localize pathological regions simultaneously. The algorithm is also applied on images from Alzheimer's disease patients to demonstrate the generality of the method.

AB - We present a generic method for automatic detection of abnormal regions in medical images as deviations from a normative data base. The algorithm decomposes an image, or more broadly a function defined on the image grid, into the superposition of a normal part and a residual term. A statistical model is constructed with regional sparse learning to represent normative anatomical variations among a reference population (e.g., healthy controls), in conjunction with a Markov random field regularization that ensures mutual consistency of the regional learning among partially overlapping image blocks. The decomposition is performed in a principled way so that the normal part fits well with the learned normative model, while the residual term absorbs pathological patterns, which may then be detected through a statistical significance test. The decomposition is applied to multiple image features from an individual scan, detecting abnormalities using both intensity and shape information. We form an iterative scheme that interleaves abnormality detection with deformable registration, gradually improving robustness of the spatial normalization and precision of the detection. The algorithm is evaluated with simulated images and clinical data of brain lesions, and is shown to achieve robust deformable registration and localize pathological regions simultaneously. The algorithm is also applied on images from Alzheimer's disease patients to demonstrate the generality of the method.

KW - Abnormality detection

KW - basis pursuit

KW - brain pathology

KW - convex optimization

KW - medical image registration

KW - sparse representation

UR - http://www.scopus.com/inward/record.url?scp=84986203192&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84986203192&partnerID=8YFLogxK

U2 - 10.1109/TMI.2016.2538998

DO - 10.1109/TMI.2016.2538998

M3 - Article

VL - 35

SP - 1937

EP - 1951

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 8

M1 - 7426805

ER -