Please Don't Move—Evaluating Motion Artifact From Peripheral Quantitative Computed Tomography Scans Using Textural Features

Timo Rantalainen, Paola Chivers, Belinda R. Beck, Sam Robertson, Nicolas H. Hart, Sophia Nimphius, Benjamin K. Weeks, Fleur McIntyre, Beth Hands, Aris Siafarikas

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Most imaging methods, including peripheral quantitative computed tomography (pQCT), are susceptible to motion artifacts particularly in fidgety pediatric populations. Methods currently used to address motion artifact include manual screening (visual inspection) and objective assessments of the scans. However, previously reported objective methods either cannot be applied on the reconstructed image or have not been tested for distal bone sites. Therefore, the purpose of the present study was to develop and validate motion artifact classifiers to quantify motion artifact in pQCT scans. Whether textural features could provide adequate motion artifact classification performance in 2 adolescent datasets with pQCT scans from tibial and radial diaphyses and epiphyses was tested. The first dataset was split into training (66% of sample) and validation (33% of sample) datasets. Visual classification was used as the ground truth. Moderate to substantial classification performance (J48 classifier, kappa coefficients from 0.57 to 0.80) was observed in the validation dataset with the novel texture-based classifier. In applying the same classifier to the second cross-sectional dataset, a slight-to-fair (κ = 0.01–0.39) classification performance was observed. Overall, this novel textural analysis-based classifier provided a moderate-to-substantial classification of motion artifact when the classifier was specifically trained for the measurement device and population. Classification based on textural features may be used to prescreen obviously acceptable and unacceptable scans, with a subsequent human-operated visual classification of any remaining scans.

Original languageEnglish
Pages (from-to)260-268
Number of pages9
JournalJournal of Clinical Densitometry
Volume21
Issue number2
DOIs
Publication statusPublished - 1 Apr 2018

Fingerprint

Artifacts
Tomography
Diaphyses
Epiphyses
Population
Datasets
Pediatrics
Bone and Bones
Equipment and Supplies

Cite this

Rantalainen, Timo ; Chivers, Paola ; Beck, Belinda R. ; Robertson, Sam ; Hart, Nicolas H. ; Nimphius, Sophia ; Weeks, Benjamin K. ; McIntyre, Fleur ; Hands, Beth ; Siafarikas, Aris. / Please Don't Move—Evaluating Motion Artifact From Peripheral Quantitative Computed Tomography Scans Using Textural Features. In: Journal of Clinical Densitometry. 2018 ; Vol. 21, No. 2. pp. 260-268.
@article{420408d441f3486eb3f790449329fa53,
title = "Please Don't Move—Evaluating Motion Artifact From Peripheral Quantitative Computed Tomography Scans Using Textural Features",
abstract = "Most imaging methods, including peripheral quantitative computed tomography (pQCT), are susceptible to motion artifacts particularly in fidgety pediatric populations. Methods currently used to address motion artifact include manual screening (visual inspection) and objective assessments of the scans. However, previously reported objective methods either cannot be applied on the reconstructed image or have not been tested for distal bone sites. Therefore, the purpose of the present study was to develop and validate motion artifact classifiers to quantify motion artifact in pQCT scans. Whether textural features could provide adequate motion artifact classification performance in 2 adolescent datasets with pQCT scans from tibial and radial diaphyses and epiphyses was tested. The first dataset was split into training (66{\%} of sample) and validation (33{\%} of sample) datasets. Visual classification was used as the ground truth. Moderate to substantial classification performance (J48 classifier, kappa coefficients from 0.57 to 0.80) was observed in the validation dataset with the novel texture-based classifier. In applying the same classifier to the second cross-sectional dataset, a slight-to-fair (κ = 0.01–0.39) classification performance was observed. Overall, this novel textural analysis-based classifier provided a moderate-to-substantial classification of motion artifact when the classifier was specifically trained for the measurement device and population. Classification based on textural features may be used to prescreen obviously acceptable and unacceptable scans, with a subsequent human-operated visual classification of any remaining scans.",
keywords = "Bone QCT, machine learning, morphology, precision, repeatability",
author = "Timo Rantalainen and Paola Chivers and Beck, {Belinda R.} and Sam Robertson and Hart, {Nicolas H.} and Sophia Nimphius and Weeks, {Benjamin K.} and Fleur McIntyre and Beth Hands and Aris Siafarikas",
year = "2018",
month = "4",
day = "1",
doi = "10.1016/j.jocd.2017.07.002",
language = "English",
volume = "21",
pages = "260--268",
journal = "Journal of Clinical Densitometry",
issn = "1094-6950",
publisher = "Elsevier",
number = "2",

}

Rantalainen, T, Chivers, P, Beck, BR, Robertson, S, Hart, NH, Nimphius, S, Weeks, BK, McIntyre, F, Hands, B & Siafarikas, A 2018, 'Please Don't Move—Evaluating Motion Artifact From Peripheral Quantitative Computed Tomography Scans Using Textural Features' Journal of Clinical Densitometry, vol. 21, no. 2, pp. 260-268. https://doi.org/10.1016/j.jocd.2017.07.002

Please Don't Move—Evaluating Motion Artifact From Peripheral Quantitative Computed Tomography Scans Using Textural Features. / Rantalainen, Timo; Chivers, Paola; Beck, Belinda R.; Robertson, Sam; Hart, Nicolas H.; Nimphius, Sophia; Weeks, Benjamin K.; McIntyre, Fleur; Hands, Beth; Siafarikas, Aris.

In: Journal of Clinical Densitometry, Vol. 21, No. 2, 01.04.2018, p. 260-268.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Please Don't Move—Evaluating Motion Artifact From Peripheral Quantitative Computed Tomography Scans Using Textural Features

AU - Rantalainen, Timo

AU - Chivers, Paola

AU - Beck, Belinda R.

AU - Robertson, Sam

AU - Hart, Nicolas H.

AU - Nimphius, Sophia

AU - Weeks, Benjamin K.

AU - McIntyre, Fleur

AU - Hands, Beth

AU - Siafarikas, Aris

PY - 2018/4/1

Y1 - 2018/4/1

N2 - Most imaging methods, including peripheral quantitative computed tomography (pQCT), are susceptible to motion artifacts particularly in fidgety pediatric populations. Methods currently used to address motion artifact include manual screening (visual inspection) and objective assessments of the scans. However, previously reported objective methods either cannot be applied on the reconstructed image or have not been tested for distal bone sites. Therefore, the purpose of the present study was to develop and validate motion artifact classifiers to quantify motion artifact in pQCT scans. Whether textural features could provide adequate motion artifact classification performance in 2 adolescent datasets with pQCT scans from tibial and radial diaphyses and epiphyses was tested. The first dataset was split into training (66% of sample) and validation (33% of sample) datasets. Visual classification was used as the ground truth. Moderate to substantial classification performance (J48 classifier, kappa coefficients from 0.57 to 0.80) was observed in the validation dataset with the novel texture-based classifier. In applying the same classifier to the second cross-sectional dataset, a slight-to-fair (κ = 0.01–0.39) classification performance was observed. Overall, this novel textural analysis-based classifier provided a moderate-to-substantial classification of motion artifact when the classifier was specifically trained for the measurement device and population. Classification based on textural features may be used to prescreen obviously acceptable and unacceptable scans, with a subsequent human-operated visual classification of any remaining scans.

AB - Most imaging methods, including peripheral quantitative computed tomography (pQCT), are susceptible to motion artifacts particularly in fidgety pediatric populations. Methods currently used to address motion artifact include manual screening (visual inspection) and objective assessments of the scans. However, previously reported objective methods either cannot be applied on the reconstructed image or have not been tested for distal bone sites. Therefore, the purpose of the present study was to develop and validate motion artifact classifiers to quantify motion artifact in pQCT scans. Whether textural features could provide adequate motion artifact classification performance in 2 adolescent datasets with pQCT scans from tibial and radial diaphyses and epiphyses was tested. The first dataset was split into training (66% of sample) and validation (33% of sample) datasets. Visual classification was used as the ground truth. Moderate to substantial classification performance (J48 classifier, kappa coefficients from 0.57 to 0.80) was observed in the validation dataset with the novel texture-based classifier. In applying the same classifier to the second cross-sectional dataset, a slight-to-fair (κ = 0.01–0.39) classification performance was observed. Overall, this novel textural analysis-based classifier provided a moderate-to-substantial classification of motion artifact when the classifier was specifically trained for the measurement device and population. Classification based on textural features may be used to prescreen obviously acceptable and unacceptable scans, with a subsequent human-operated visual classification of any remaining scans.

KW - Bone QCT

KW - machine learning

KW - morphology

KW - precision

KW - repeatability

UR - http://www.scopus.com/inward/record.url?scp=85027184307&partnerID=8YFLogxK

U2 - 10.1016/j.jocd.2017.07.002

DO - 10.1016/j.jocd.2017.07.002

M3 - Article

VL - 21

SP - 260

EP - 268

JO - Journal of Clinical Densitometry

JF - Journal of Clinical Densitometry

SN - 1094-6950

IS - 2

ER -