Automatic analysis of trabecular bone structure from knee MRI

Joselene Marques; Rabia Granlund; Martin Lillholm; Paola C. Pettersen; Erik B.  Dam

doi:10.1016/j.compbiomed.2012.04.005

Automatic analysis of trabecular bone structure from knee MRI

Joselene Marques, Rabia Granlund, Martin Lillholm, Paola C. Pettersen, Erik B. Dam

Department of Computer Science

4 Citations (Scopus)

Abstract

We investigated the feasibility of quantifying osteoarthritis (OA) by analysis of the trabecular bone structure in low-field knee MRI. Generic texture features were extracted from the images and subsequently selected by sequential floating forward selection (SFFS), following a fully automatic, uncommitted machine-learning based framework. Six different classifiers were evaluated in cross-validation schemes and the results showed that the presence of OA can be quantified by a bone structure marker. The performance of the developed marker reached a generalization area-under-the-ROC (AUC) of 0.82, which is higher than the established cartilage markers known to relate to the OA diagnosis.

Original language	English
Journal	Computers in Biology and Medicine
Volume	42
Issue number	7
Pages (from-to)	735-742
Number of pages	8
ISSN	0010-4825
DOIs	https://doi.org/10.1016/j.compbiomed.2012.04.005
Publication status	Published - Jul 2012

Access to Document

10.1016/j.compbiomed.2012.04.005

Automatic analysis of trabecular bone structure from knee MRIFinal published version, 352 KB

Cite this

@article{e6608ae37bb4424abd28614233d3d4c8,

title = "Automatic analysis of trabecular bone structure from knee MRI",

abstract = "We investigated the feasibility of quantifying osteoarthritis (OA) by analysis of the trabecular bone structure in low-field knee MRI. Generic texture features were extracted from the images and subsequently selected by sequential floating forward selection (SFFS), following a fully automatic, uncommitted machine-learning based framework. Six different classifiers were evaluated in cross-validation schemes and the results showed that the presence of OA can be quantified by a bone structure marker. The performance of the developed marker reached a generalization area-under-the-ROC (AUC) of 0.82, which is higher than the established cartilage markers known to relate to the OA diagnosis.",

author = "Joselene Marques and Rabia Granlund and Martin Lillholm and Pettersen, {Paola C.} and Dam, {Erik B.}",

year = "2012",

month = jul,

doi = "10.1016/j.compbiomed.2012.04.005",

language = "English",

volume = "42",

pages = "735--742",

journal = "Computers in Biology and Medicine",

issn = "0010-4825",

publisher = "Pergamon Press",

number = "7",

}

TY - JOUR

T1 - Automatic analysis of trabecular bone structure from knee MRI

AU - Marques, Joselene

AU - Granlund, Rabia

AU - Lillholm, Martin

AU - Pettersen, Paola C.

AU - Dam, Erik B.

PY - 2012/7

Y1 - 2012/7

N2 - We investigated the feasibility of quantifying osteoarthritis (OA) by analysis of the trabecular bone structure in low-field knee MRI. Generic texture features were extracted from the images and subsequently selected by sequential floating forward selection (SFFS), following a fully automatic, uncommitted machine-learning based framework. Six different classifiers were evaluated in cross-validation schemes and the results showed that the presence of OA can be quantified by a bone structure marker. The performance of the developed marker reached a generalization area-under-the-ROC (AUC) of 0.82, which is higher than the established cartilage markers known to relate to the OA diagnosis.

AB - We investigated the feasibility of quantifying osteoarthritis (OA) by analysis of the trabecular bone structure in low-field knee MRI. Generic texture features were extracted from the images and subsequently selected by sequential floating forward selection (SFFS), following a fully automatic, uncommitted machine-learning based framework. Six different classifiers were evaluated in cross-validation schemes and the results showed that the presence of OA can be quantified by a bone structure marker. The performance of the developed marker reached a generalization area-under-the-ROC (AUC) of 0.82, which is higher than the established cartilage markers known to relate to the OA diagnosis.

U2 - 10.1016/j.compbiomed.2012.04.005

DO - 10.1016/j.compbiomed.2012.04.005

M3 - Journal article

C2 - 22579046

SN - 0010-4825

VL - 42

SP - 735

EP - 742

JO - Computers in Biology and Medicine

JF - Computers in Biology and Medicine

IS - 7

ER -

Automatic analysis of trabecular bone structure from knee MRI

Abstract

Access to Document

Fingerprint

Cite this