Can we improve the prediction of hip fracture by assessing bone structure using shape and appearance modelling?

S. R. Goodyear, R. J. Barr, E. McCloskey, S. Alesci, R. M. Aspden, D. M. Reid, J. S. Gregory

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

Purpose: There is a continuing need to improve the prediction of hip fractures to identify those at highest risk, enabling cost-effective use of preventative therapies. Methods: The aim of this work was to validate an innovative imaging biomarker for hip fracture by modelling the shape and texture of the proximal femur assessed from dual energy X-ray absorptiometry (DXA) scans. Scans used were acquired at baseline from elderly patients participating in a prospective, placebo-controlled fracture prevention study of the bisphosphonate, clodronate. 182 subjects who subsequently suffered a hip fracture were age, weight and height matched with two controls who did not suffer a fracture during a median 4-year follow-up period. Logistic regression was used to test if variables were good predictors of fracture and adjust for bone mineral density (BMD). Results: Shape mode 2, reflecting variability in neck-shaft angle, neck width and the size of both trochanters (0.81 (OR), 0.68-0.97 (CI), 0.024 (P)), and appearance mode 6, recording grey-level contrast (1.33, 1.11-1.59, 0.002), were significant predictors of hip fracture and remained so after adjustment for BMD (shape mode 2 (0.77, 0.64-0.93, 0.006), appearance mode 6 (1.32, 1.10-1.59, 0.003)). Receiver Operating Curve analysis showed the combination of shape mode 2, appearance mode 6 and BMD was 3% better than any single predictor. Conclusion: Variables derived from shape and appearance models gave a prediction of fracture comparable to BMD and in combination with BMD gave an improvement in the prediction of hip fracture that could predict an additional 2000 hip fracture cases per year in the UK, potentially saving more than £20 million per year and 10,000 cases in the US.
Original languageEnglish
Pages (from-to)188-193
Number of pages6
JournalBone
Volume53
Issue number1
DOIs
Publication statusPublished - 1 Mar 2013

Fingerprint

Hip Fractures
Bone Density
Bone and Bones
Femur
Neck
Clodronic Acid
Photon Absorptiometry
Diphosphonates
Biomarkers
Logistic Models
Placebos
Weights and Measures
Costs and Cost Analysis

Keywords

  • prediction
  • geometric factors
  • osteoperosis
  • hip fracture
  • active shape modelling
  • statistical-model
  • X-Ray absorptiometry
  • axis length
  • proximal femur
  • mineral density
  • women
  • femoral-neck
  • epidemiology

Cite this

Can we improve the prediction of hip fracture by assessing bone structure using shape and appearance modelling? / Goodyear, S. R.; Barr, R. J.; McCloskey, E.; Alesci, S.; Aspden, R. M.; Reid, D. M.; Gregory, J. S. .

In: Bone, Vol. 53, No. 1, 01.03.2013, p. 188-193.

Research output: Contribution to journalArticle

Goodyear, S. R. ; Barr, R. J. ; McCloskey, E. ; Alesci, S. ; Aspden, R. M. ; Reid, D. M. ; Gregory, J. S. . / Can we improve the prediction of hip fracture by assessing bone structure using shape and appearance modelling?. In: Bone. 2013 ; Vol. 53, No. 1. pp. 188-193.
@article{48529ae3d37348dd8ea008f24183d7ca,
title = "Can we improve the prediction of hip fracture by assessing bone structure using shape and appearance modelling?",
abstract = "Purpose: There is a continuing need to improve the prediction of hip fractures to identify those at highest risk, enabling cost-effective use of preventative therapies. Methods: The aim of this work was to validate an innovative imaging biomarker for hip fracture by modelling the shape and texture of the proximal femur assessed from dual energy X-ray absorptiometry (DXA) scans. Scans used were acquired at baseline from elderly patients participating in a prospective, placebo-controlled fracture prevention study of the bisphosphonate, clodronate. 182 subjects who subsequently suffered a hip fracture were age, weight and height matched with two controls who did not suffer a fracture during a median 4-year follow-up period. Logistic regression was used to test if variables were good predictors of fracture and adjust for bone mineral density (BMD). Results: Shape mode 2, reflecting variability in neck-shaft angle, neck width and the size of both trochanters (0.81 (OR), 0.68-0.97 (CI), 0.024 (P)), and appearance mode 6, recording grey-level contrast (1.33, 1.11-1.59, 0.002), were significant predictors of hip fracture and remained so after adjustment for BMD (shape mode 2 (0.77, 0.64-0.93, 0.006), appearance mode 6 (1.32, 1.10-1.59, 0.003)). Receiver Operating Curve analysis showed the combination of shape mode 2, appearance mode 6 and BMD was 3{\%} better than any single predictor. Conclusion: Variables derived from shape and appearance models gave a prediction of fracture comparable to BMD and in combination with BMD gave an improvement in the prediction of hip fracture that could predict an additional 2000 hip fracture cases per year in the UK, potentially saving more than £20 million per year and 10,000 cases in the US.",
keywords = "prediction, geometric factors, osteoperosis, hip fracture, active shape modelling, statistical-model, X-Ray absorptiometry, axis length, proximal femur, mineral density, women, femoral-neck, epidemiology",
author = "Goodyear, {S. R.} and Barr, {R. J.} and E. McCloskey and S. Alesci and Aspden, {R. M.} and Reid, {D. M.} and Gregory, {J. S.}",
note = "Copyright 2013 Elsevier B.V., All rights reserved.",
year = "2013",
month = "3",
day = "1",
doi = "10.1016/j.bone.2012.11.042",
language = "English",
volume = "53",
pages = "188--193",
journal = "Bone",
issn = "8756-3282",
publisher = "Elsevier Inc.",
number = "1",

}

TY - JOUR

T1 - Can we improve the prediction of hip fracture by assessing bone structure using shape and appearance modelling?

AU - Goodyear, S. R.

AU - Barr, R. J.

AU - McCloskey, E.

AU - Alesci, S.

AU - Aspden, R. M.

AU - Reid, D. M.

AU - Gregory, J. S.

N1 - Copyright 2013 Elsevier B.V., All rights reserved.

PY - 2013/3/1

Y1 - 2013/3/1

N2 - Purpose: There is a continuing need to improve the prediction of hip fractures to identify those at highest risk, enabling cost-effective use of preventative therapies. Methods: The aim of this work was to validate an innovative imaging biomarker for hip fracture by modelling the shape and texture of the proximal femur assessed from dual energy X-ray absorptiometry (DXA) scans. Scans used were acquired at baseline from elderly patients participating in a prospective, placebo-controlled fracture prevention study of the bisphosphonate, clodronate. 182 subjects who subsequently suffered a hip fracture were age, weight and height matched with two controls who did not suffer a fracture during a median 4-year follow-up period. Logistic regression was used to test if variables were good predictors of fracture and adjust for bone mineral density (BMD). Results: Shape mode 2, reflecting variability in neck-shaft angle, neck width and the size of both trochanters (0.81 (OR), 0.68-0.97 (CI), 0.024 (P)), and appearance mode 6, recording grey-level contrast (1.33, 1.11-1.59, 0.002), were significant predictors of hip fracture and remained so after adjustment for BMD (shape mode 2 (0.77, 0.64-0.93, 0.006), appearance mode 6 (1.32, 1.10-1.59, 0.003)). Receiver Operating Curve analysis showed the combination of shape mode 2, appearance mode 6 and BMD was 3% better than any single predictor. Conclusion: Variables derived from shape and appearance models gave a prediction of fracture comparable to BMD and in combination with BMD gave an improvement in the prediction of hip fracture that could predict an additional 2000 hip fracture cases per year in the UK, potentially saving more than £20 million per year and 10,000 cases in the US.

AB - Purpose: There is a continuing need to improve the prediction of hip fractures to identify those at highest risk, enabling cost-effective use of preventative therapies. Methods: The aim of this work was to validate an innovative imaging biomarker for hip fracture by modelling the shape and texture of the proximal femur assessed from dual energy X-ray absorptiometry (DXA) scans. Scans used were acquired at baseline from elderly patients participating in a prospective, placebo-controlled fracture prevention study of the bisphosphonate, clodronate. 182 subjects who subsequently suffered a hip fracture were age, weight and height matched with two controls who did not suffer a fracture during a median 4-year follow-up period. Logistic regression was used to test if variables were good predictors of fracture and adjust for bone mineral density (BMD). Results: Shape mode 2, reflecting variability in neck-shaft angle, neck width and the size of both trochanters (0.81 (OR), 0.68-0.97 (CI), 0.024 (P)), and appearance mode 6, recording grey-level contrast (1.33, 1.11-1.59, 0.002), were significant predictors of hip fracture and remained so after adjustment for BMD (shape mode 2 (0.77, 0.64-0.93, 0.006), appearance mode 6 (1.32, 1.10-1.59, 0.003)). Receiver Operating Curve analysis showed the combination of shape mode 2, appearance mode 6 and BMD was 3% better than any single predictor. Conclusion: Variables derived from shape and appearance models gave a prediction of fracture comparable to BMD and in combination with BMD gave an improvement in the prediction of hip fracture that could predict an additional 2000 hip fracture cases per year in the UK, potentially saving more than £20 million per year and 10,000 cases in the US.

KW - prediction

KW - geometric factors

KW - osteoperosis

KW - hip fracture

KW - active shape modelling

KW - statistical-model

KW - X-Ray absorptiometry

KW - axis length

KW - proximal femur

KW - mineral density

KW - women

KW - femoral-neck

KW - epidemiology

UR - http://www.scopus.com/inward/record.url?scp=84871998823&partnerID=8YFLogxK

U2 - 10.1016/j.bone.2012.11.042

DO - 10.1016/j.bone.2012.11.042

M3 - Article

VL - 53

SP - 188

EP - 193

JO - Bone

JF - Bone

SN - 8756-3282

IS - 1

ER -