Computer-assisted measurement of the size of ossification in patients with ossification of the posterior longitudinal ligament in the cervical spine

Kazuhiro Chiba, Yoshiharu Kato, Nobuyuki Tsuzuki, Kensei Nagata, Yoshiaki Toyama, Motoki Iwasaki, Kazuo Yonenobu

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Background. Progression of ossification of the posterior longitudinal ligament in patients may lead to serious neurological deterioration. A government-funded study group established a manual method of measurement on plain radiographs to detect progression of the ossified lesion. However, this method did not gain wide acceptance because it was time-consuming and complicated, for which drawings of many lines and points are required. We have applied a computer-assisted measurement system to this task and have evaluated inter- and intraexaminer reliability, showing that it is quicker to use and more accurate than the manual method. Methods. Eight board-certified spine surgeons, acting as the examiners, measured the sizes of the ossified lesions on nine lateral cervical spine radiographs using the computer-assisted measurement system. Following insertion of digitized radiographic image data into a computer, the corners of the vertebral bodies on the displayed images are marked by the examiners, and the software automatically sets reference lines and points. The examiners identify upper, lower, and posterior margins of the ossified lesions, and the software calculates the dimensions of the ossified lesions. Data obtained from eight examiners for length and thickness underwent rigorous statistical analysis by calculating the intraclass correlation coefficients with 95% confidence intervals (CIs) to determine interexaminer reliability and Pearson's correlation coefficients between the two measurements by the same examiner to determine intraexaminer reliability. Results. The intraclass correlation coefficients were 0.927 and 0.968 with 95% CIs of 0.883-0.955 and 0.956-0.978 for measurements of length and thickness, respectively, of the ossified lesions. The Pearson's correlation coefficients for the two measurements by the same examiners were 0.943-0.985 for length and 0.957-0.991 for thickness. Conclusions. The inter- and intraexaminer reliability using this measurement system was excellent. The method can detect progression of ossification of the posterior longitudinal ligament (OPLL) on plain radiographs with high precision and could become a standard method for measuring the size of OPLL.

Original languageEnglish
Pages (from-to)451-456
Number of pages6
JournalJournal of Orthopaedic Science
Volume10
Issue number5
DOIs
Publication statusPublished - 2005 Sep

Fingerprint

Ossification of Posterior Longitudinal Ligament
Osteogenesis
Spine
Software
Confidence Intervals

ASJC Scopus subject areas

  • Surgery

Cite this

Computer-assisted measurement of the size of ossification in patients with ossification of the posterior longitudinal ligament in the cervical spine. / Chiba, Kazuhiro; Kato, Yoshiharu; Tsuzuki, Nobuyuki; Nagata, Kensei; Toyama, Yoshiaki; Iwasaki, Motoki; Yonenobu, Kazuo.

In: Journal of Orthopaedic Science, Vol. 10, No. 5, 09.2005, p. 451-456.

Research output: Contribution to journalArticle

Chiba, Kazuhiro ; Kato, Yoshiharu ; Tsuzuki, Nobuyuki ; Nagata, Kensei ; Toyama, Yoshiaki ; Iwasaki, Motoki ; Yonenobu, Kazuo. / Computer-assisted measurement of the size of ossification in patients with ossification of the posterior longitudinal ligament in the cervical spine. In: Journal of Orthopaedic Science. 2005 ; Vol. 10, No. 5. pp. 451-456.
@article{44dab0feda6a401ab8eaa3aa7b211f0c,
title = "Computer-assisted measurement of the size of ossification in patients with ossification of the posterior longitudinal ligament in the cervical spine",
abstract = "Background. Progression of ossification of the posterior longitudinal ligament in patients may lead to serious neurological deterioration. A government-funded study group established a manual method of measurement on plain radiographs to detect progression of the ossified lesion. However, this method did not gain wide acceptance because it was time-consuming and complicated, for which drawings of many lines and points are required. We have applied a computer-assisted measurement system to this task and have evaluated inter- and intraexaminer reliability, showing that it is quicker to use and more accurate than the manual method. Methods. Eight board-certified spine surgeons, acting as the examiners, measured the sizes of the ossified lesions on nine lateral cervical spine radiographs using the computer-assisted measurement system. Following insertion of digitized radiographic image data into a computer, the corners of the vertebral bodies on the displayed images are marked by the examiners, and the software automatically sets reference lines and points. The examiners identify upper, lower, and posterior margins of the ossified lesions, and the software calculates the dimensions of the ossified lesions. Data obtained from eight examiners for length and thickness underwent rigorous statistical analysis by calculating the intraclass correlation coefficients with 95{\%} confidence intervals (CIs) to determine interexaminer reliability and Pearson's correlation coefficients between the two measurements by the same examiner to determine intraexaminer reliability. Results. The intraclass correlation coefficients were 0.927 and 0.968 with 95{\%} CIs of 0.883-0.955 and 0.956-0.978 for measurements of length and thickness, respectively, of the ossified lesions. The Pearson's correlation coefficients for the two measurements by the same examiners were 0.943-0.985 for length and 0.957-0.991 for thickness. Conclusions. The inter- and intraexaminer reliability using this measurement system was excellent. The method can detect progression of ossification of the posterior longitudinal ligament (OPLL) on plain radiographs with high precision and could become a standard method for measuring the size of OPLL.",
author = "Kazuhiro Chiba and Yoshiharu Kato and Nobuyuki Tsuzuki and Kensei Nagata and Yoshiaki Toyama and Motoki Iwasaki and Kazuo Yonenobu",
year = "2005",
month = "9",
doi = "10.1007/s00776-005-0925-5",
language = "English",
volume = "10",
pages = "451--456",
journal = "Journal of Orthopaedic Science",
issn = "0949-2658",
publisher = "Springer Japan",
number = "5",

}

TY - JOUR

T1 - Computer-assisted measurement of the size of ossification in patients with ossification of the posterior longitudinal ligament in the cervical spine

AU - Chiba, Kazuhiro

AU - Kato, Yoshiharu

AU - Tsuzuki, Nobuyuki

AU - Nagata, Kensei

AU - Toyama, Yoshiaki

AU - Iwasaki, Motoki

AU - Yonenobu, Kazuo

PY - 2005/9

Y1 - 2005/9

N2 - Background. Progression of ossification of the posterior longitudinal ligament in patients may lead to serious neurological deterioration. A government-funded study group established a manual method of measurement on plain radiographs to detect progression of the ossified lesion. However, this method did not gain wide acceptance because it was time-consuming and complicated, for which drawings of many lines and points are required. We have applied a computer-assisted measurement system to this task and have evaluated inter- and intraexaminer reliability, showing that it is quicker to use and more accurate than the manual method. Methods. Eight board-certified spine surgeons, acting as the examiners, measured the sizes of the ossified lesions on nine lateral cervical spine radiographs using the computer-assisted measurement system. Following insertion of digitized radiographic image data into a computer, the corners of the vertebral bodies on the displayed images are marked by the examiners, and the software automatically sets reference lines and points. The examiners identify upper, lower, and posterior margins of the ossified lesions, and the software calculates the dimensions of the ossified lesions. Data obtained from eight examiners for length and thickness underwent rigorous statistical analysis by calculating the intraclass correlation coefficients with 95% confidence intervals (CIs) to determine interexaminer reliability and Pearson's correlation coefficients between the two measurements by the same examiner to determine intraexaminer reliability. Results. The intraclass correlation coefficients were 0.927 and 0.968 with 95% CIs of 0.883-0.955 and 0.956-0.978 for measurements of length and thickness, respectively, of the ossified lesions. The Pearson's correlation coefficients for the two measurements by the same examiners were 0.943-0.985 for length and 0.957-0.991 for thickness. Conclusions. The inter- and intraexaminer reliability using this measurement system was excellent. The method can detect progression of ossification of the posterior longitudinal ligament (OPLL) on plain radiographs with high precision and could become a standard method for measuring the size of OPLL.

AB - Background. Progression of ossification of the posterior longitudinal ligament in patients may lead to serious neurological deterioration. A government-funded study group established a manual method of measurement on plain radiographs to detect progression of the ossified lesion. However, this method did not gain wide acceptance because it was time-consuming and complicated, for which drawings of many lines and points are required. We have applied a computer-assisted measurement system to this task and have evaluated inter- and intraexaminer reliability, showing that it is quicker to use and more accurate than the manual method. Methods. Eight board-certified spine surgeons, acting as the examiners, measured the sizes of the ossified lesions on nine lateral cervical spine radiographs using the computer-assisted measurement system. Following insertion of digitized radiographic image data into a computer, the corners of the vertebral bodies on the displayed images are marked by the examiners, and the software automatically sets reference lines and points. The examiners identify upper, lower, and posterior margins of the ossified lesions, and the software calculates the dimensions of the ossified lesions. Data obtained from eight examiners for length and thickness underwent rigorous statistical analysis by calculating the intraclass correlation coefficients with 95% confidence intervals (CIs) to determine interexaminer reliability and Pearson's correlation coefficients between the two measurements by the same examiner to determine intraexaminer reliability. Results. The intraclass correlation coefficients were 0.927 and 0.968 with 95% CIs of 0.883-0.955 and 0.956-0.978 for measurements of length and thickness, respectively, of the ossified lesions. The Pearson's correlation coefficients for the two measurements by the same examiners were 0.943-0.985 for length and 0.957-0.991 for thickness. Conclusions. The inter- and intraexaminer reliability using this measurement system was excellent. The method can detect progression of ossification of the posterior longitudinal ligament (OPLL) on plain radiographs with high precision and could become a standard method for measuring the size of OPLL.

UR - http://www.scopus.com/inward/record.url?scp=26244446244&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=26244446244&partnerID=8YFLogxK

U2 - 10.1007/s00776-005-0925-5

DO - 10.1007/s00776-005-0925-5

M3 - Article

VL - 10

SP - 451

EP - 456

JO - Journal of Orthopaedic Science

JF - Journal of Orthopaedic Science

SN - 0949-2658

IS - 5

ER -