Fully automated radiogrammetric measurement of third metacarpal bone from hand radiograph

Anu Shaju Areeckal, S. Sumam David, Michel Kocher, Nikil Jayasheelan, Jagannath Kamath

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Osteoporosis is a disease caused by reduction of bone mass, bone strength and deterioration of bone structure. The gold standard method for diagnosis of osteoporosis is measurement of bone mineral density (BMD) using Dual X-ray Absorptiometry (DXA). However, DXA is expensive and not widely available in developing countries. An alternative cost-effective method for measurement of bone loss and strength is metacarpal radiogrammetry, by which geometric measurements of cortical bone of the metacarpal bone are measured. In this paper, we propose a fully automated method for segmentation of third metacarpal bone from hand radiograph and radiogrammetric measurements using mathematical morphology. Cortical width and thickness are measured from the endosteal and periosteal edges of the metacarpal bone using which bone indices which help in diagnosis of osteoporosis can be computed. The proposed segmentation method was tested on 157 hand X-ray images. A success rate of 94.9% is obtained for automatic detection of third metacarpal bone. Evaluation of cortical measurements of 3 calibrated images is done by comparing the results with ground truth. The mean accuracy error obtained was 0.02cm and 0.04cm for cortical width and medullary width, respectively.

Original languageEnglish
Title of host publication2016 International Conference on Signal Processing and Communications, SPCOM 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509017461
DOIs
Publication statusPublished - 16-11-2016
Event11th International Conference on Signal Processing and Communications, SPCOM 2016 - Bangalore, India
Duration: 12-06-201615-06-2016

Conference

Conference11th International Conference on Signal Processing and Communications, SPCOM 2016
CountryIndia
CityBangalore
Period12-06-1615-06-16

Fingerprint

Bone
X rays
Mathematical morphology
Developing countries
Deterioration
Minerals

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Signal Processing

Cite this

Areeckal, A. S., Sumam David, S., Kocher, M., Jayasheelan, N., & Kamath, J. (2016). Fully automated radiogrammetric measurement of third metacarpal bone from hand radiograph. In 2016 International Conference on Signal Processing and Communications, SPCOM 2016 [7746608] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SPCOM.2016.7746608
Areeckal, Anu Shaju ; Sumam David, S. ; Kocher, Michel ; Jayasheelan, Nikil ; Kamath, Jagannath. / Fully automated radiogrammetric measurement of third metacarpal bone from hand radiograph. 2016 International Conference on Signal Processing and Communications, SPCOM 2016. Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{988724420874496485c021ba9b78a431,
title = "Fully automated radiogrammetric measurement of third metacarpal bone from hand radiograph",
abstract = "Osteoporosis is a disease caused by reduction of bone mass, bone strength and deterioration of bone structure. The gold standard method for diagnosis of osteoporosis is measurement of bone mineral density (BMD) using Dual X-ray Absorptiometry (DXA). However, DXA is expensive and not widely available in developing countries. An alternative cost-effective method for measurement of bone loss and strength is metacarpal radiogrammetry, by which geometric measurements of cortical bone of the metacarpal bone are measured. In this paper, we propose a fully automated method for segmentation of third metacarpal bone from hand radiograph and radiogrammetric measurements using mathematical morphology. Cortical width and thickness are measured from the endosteal and periosteal edges of the metacarpal bone using which bone indices which help in diagnosis of osteoporosis can be computed. The proposed segmentation method was tested on 157 hand X-ray images. A success rate of 94.9{\%} is obtained for automatic detection of third metacarpal bone. Evaluation of cortical measurements of 3 calibrated images is done by comparing the results with ground truth. The mean accuracy error obtained was 0.02cm and 0.04cm for cortical width and medullary width, respectively.",
author = "Areeckal, {Anu Shaju} and {Sumam David}, S. and Michel Kocher and Nikil Jayasheelan and Jagannath Kamath",
year = "2016",
month = "11",
day = "16",
doi = "10.1109/SPCOM.2016.7746608",
language = "English",
booktitle = "2016 International Conference on Signal Processing and Communications, SPCOM 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Areeckal, AS, Sumam David, S, Kocher, M, Jayasheelan, N & Kamath, J 2016, Fully automated radiogrammetric measurement of third metacarpal bone from hand radiograph. in 2016 International Conference on Signal Processing and Communications, SPCOM 2016., 7746608, Institute of Electrical and Electronics Engineers Inc., 11th International Conference on Signal Processing and Communications, SPCOM 2016, Bangalore, India, 12-06-16. https://doi.org/10.1109/SPCOM.2016.7746608

Fully automated radiogrammetric measurement of third metacarpal bone from hand radiograph. / Areeckal, Anu Shaju; Sumam David, S.; Kocher, Michel; Jayasheelan, Nikil; Kamath, Jagannath.

2016 International Conference on Signal Processing and Communications, SPCOM 2016. Institute of Electrical and Electronics Engineers Inc., 2016. 7746608.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Fully automated radiogrammetric measurement of third metacarpal bone from hand radiograph

AU - Areeckal, Anu Shaju

AU - Sumam David, S.

AU - Kocher, Michel

AU - Jayasheelan, Nikil

AU - Kamath, Jagannath

PY - 2016/11/16

Y1 - 2016/11/16

N2 - Osteoporosis is a disease caused by reduction of bone mass, bone strength and deterioration of bone structure. The gold standard method for diagnosis of osteoporosis is measurement of bone mineral density (BMD) using Dual X-ray Absorptiometry (DXA). However, DXA is expensive and not widely available in developing countries. An alternative cost-effective method for measurement of bone loss and strength is metacarpal radiogrammetry, by which geometric measurements of cortical bone of the metacarpal bone are measured. In this paper, we propose a fully automated method for segmentation of third metacarpal bone from hand radiograph and radiogrammetric measurements using mathematical morphology. Cortical width and thickness are measured from the endosteal and periosteal edges of the metacarpal bone using which bone indices which help in diagnosis of osteoporosis can be computed. The proposed segmentation method was tested on 157 hand X-ray images. A success rate of 94.9% is obtained for automatic detection of third metacarpal bone. Evaluation of cortical measurements of 3 calibrated images is done by comparing the results with ground truth. The mean accuracy error obtained was 0.02cm and 0.04cm for cortical width and medullary width, respectively.

AB - Osteoporosis is a disease caused by reduction of bone mass, bone strength and deterioration of bone structure. The gold standard method for diagnosis of osteoporosis is measurement of bone mineral density (BMD) using Dual X-ray Absorptiometry (DXA). However, DXA is expensive and not widely available in developing countries. An alternative cost-effective method for measurement of bone loss and strength is metacarpal radiogrammetry, by which geometric measurements of cortical bone of the metacarpal bone are measured. In this paper, we propose a fully automated method for segmentation of third metacarpal bone from hand radiograph and radiogrammetric measurements using mathematical morphology. Cortical width and thickness are measured from the endosteal and periosteal edges of the metacarpal bone using which bone indices which help in diagnosis of osteoporosis can be computed. The proposed segmentation method was tested on 157 hand X-ray images. A success rate of 94.9% is obtained for automatic detection of third metacarpal bone. Evaluation of cortical measurements of 3 calibrated images is done by comparing the results with ground truth. The mean accuracy error obtained was 0.02cm and 0.04cm for cortical width and medullary width, respectively.

UR - http://www.scopus.com/inward/record.url?scp=85003791550&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85003791550&partnerID=8YFLogxK

U2 - 10.1109/SPCOM.2016.7746608

DO - 10.1109/SPCOM.2016.7746608

M3 - Conference contribution

AN - SCOPUS:85003791550

BT - 2016 International Conference on Signal Processing and Communications, SPCOM 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Areeckal AS, Sumam David S, Kocher M, Jayasheelan N, Kamath J. Fully automated radiogrammetric measurement of third metacarpal bone from hand radiograph. In 2016 International Conference on Signal Processing and Communications, SPCOM 2016. Institute of Electrical and Electronics Engineers Inc. 2016. 7746608 https://doi.org/10.1109/SPCOM.2016.7746608