Hairline fracture detection using MRF and gibbs sampling

A. S. Chowdhury, A. Bhattacharya, S. M. Bhandarkar, G. S. Datta, Jack C Yu, Ramon E Figueroa Ortiz

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

5 Citations (Scopus)

Abstract

Detection of hairline fractures, representing points or areas of discontinuity in the bone, is a clinically challenging task, especially in presence of noise. The above problem is equally appealing from a computer vision or pattern recognition perspective since (a) traditional techniques for detection of corners, denoting points of surface discontinuity, typically fail in such cases and, (b) one needs to implicitly handle unknown local degradation in the image. A novel two-phase scheme for hairline mandibular fracture detection, that is robust to noise, is proposed. In the first phase, the hairline fractures are coarsely localized using statistical correlation and by exploiting the bilateral symmetry of the human mandible. In the second phase, the fractures are precisely identified and highlighted using a Markov Random Field (MRF) modeling approach coupled with Maximum A Posteriori probability (MAP) estimation. Gibbs sampling is used to maximize the posterior probability. Experimental results on Computer Tomography (CT) scans from real patients are presented.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007
DOIs
StatePublished - Aug 1 2007
Event7th IEEE Workshop on Applications of Computer Vision, WACV 2007 - Austin, TX, United States
Duration: Feb 21 2007Feb 22 2007

Publication series

NameProceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007

Other

Other7th IEEE Workshop on Applications of Computer Vision, WACV 2007
CountryUnited States
CityAustin, TX
Period2/21/072/22/07

Fingerprint

Sampling
Computer vision
Pattern recognition
Tomography
Bone
Degradation

Keywords

  • Computer tomography
  • Gibbs sampling
  • Hairline fractures
  • MAP
  • MRF

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Software

Cite this

Chowdhury, A. S., Bhattacharya, A., Bhandarkar, S. M., Datta, G. S., Yu, J. C., & Figueroa Ortiz, R. E. (2007). Hairline fracture detection using MRF and gibbs sampling. In Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007 [4118785] (Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007). https://doi.org/10.1109/WACV.2007.28

Hairline fracture detection using MRF and gibbs sampling. / Chowdhury, A. S.; Bhattacharya, A.; Bhandarkar, S. M.; Datta, G. S.; Yu, Jack C; Figueroa Ortiz, Ramon E.

Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007. 2007. 4118785 (Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007).

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

Chowdhury, AS, Bhattacharya, A, Bhandarkar, SM, Datta, GS, Yu, JC & Figueroa Ortiz, RE 2007, Hairline fracture detection using MRF and gibbs sampling. in Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007., 4118785, Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007, 7th IEEE Workshop on Applications of Computer Vision, WACV 2007, Austin, TX, United States, 2/21/07. https://doi.org/10.1109/WACV.2007.28
Chowdhury AS, Bhattacharya A, Bhandarkar SM, Datta GS, Yu JC, Figueroa Ortiz RE. Hairline fracture detection using MRF and gibbs sampling. In Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007. 2007. 4118785. (Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007). https://doi.org/10.1109/WACV.2007.28
Chowdhury, A. S. ; Bhattacharya, A. ; Bhandarkar, S. M. ; Datta, G. S. ; Yu, Jack C ; Figueroa Ortiz, Ramon E. / Hairline fracture detection using MRF and gibbs sampling. Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007. 2007. (Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007).
@inproceedings{71fa4d38f9ff4ef2824af7f3cd44aa80,
title = "Hairline fracture detection using MRF and gibbs sampling",
abstract = "Detection of hairline fractures, representing points or areas of discontinuity in the bone, is a clinically challenging task, especially in presence of noise. The above problem is equally appealing from a computer vision or pattern recognition perspective since (a) traditional techniques for detection of corners, denoting points of surface discontinuity, typically fail in such cases and, (b) one needs to implicitly handle unknown local degradation in the image. A novel two-phase scheme for hairline mandibular fracture detection, that is robust to noise, is proposed. In the first phase, the hairline fractures are coarsely localized using statistical correlation and by exploiting the bilateral symmetry of the human mandible. In the second phase, the fractures are precisely identified and highlighted using a Markov Random Field (MRF) modeling approach coupled with Maximum A Posteriori probability (MAP) estimation. Gibbs sampling is used to maximize the posterior probability. Experimental results on Computer Tomography (CT) scans from real patients are presented.",
keywords = "Computer tomography, Gibbs sampling, Hairline fractures, MAP, MRF",
author = "Chowdhury, {A. S.} and A. Bhattacharya and Bhandarkar, {S. M.} and Datta, {G. S.} and Yu, {Jack C} and {Figueroa Ortiz}, {Ramon E}",
year = "2007",
month = "8",
day = "1",
doi = "10.1109/WACV.2007.28",
language = "English (US)",
isbn = "0769527949",
series = "Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007",
booktitle = "Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007",

}

TY - GEN

T1 - Hairline fracture detection using MRF and gibbs sampling

AU - Chowdhury, A. S.

AU - Bhattacharya, A.

AU - Bhandarkar, S. M.

AU - Datta, G. S.

AU - Yu, Jack C

AU - Figueroa Ortiz, Ramon E

PY - 2007/8/1

Y1 - 2007/8/1

N2 - Detection of hairline fractures, representing points or areas of discontinuity in the bone, is a clinically challenging task, especially in presence of noise. The above problem is equally appealing from a computer vision or pattern recognition perspective since (a) traditional techniques for detection of corners, denoting points of surface discontinuity, typically fail in such cases and, (b) one needs to implicitly handle unknown local degradation in the image. A novel two-phase scheme for hairline mandibular fracture detection, that is robust to noise, is proposed. In the first phase, the hairline fractures are coarsely localized using statistical correlation and by exploiting the bilateral symmetry of the human mandible. In the second phase, the fractures are precisely identified and highlighted using a Markov Random Field (MRF) modeling approach coupled with Maximum A Posteriori probability (MAP) estimation. Gibbs sampling is used to maximize the posterior probability. Experimental results on Computer Tomography (CT) scans from real patients are presented.

AB - Detection of hairline fractures, representing points or areas of discontinuity in the bone, is a clinically challenging task, especially in presence of noise. The above problem is equally appealing from a computer vision or pattern recognition perspective since (a) traditional techniques for detection of corners, denoting points of surface discontinuity, typically fail in such cases and, (b) one needs to implicitly handle unknown local degradation in the image. A novel two-phase scheme for hairline mandibular fracture detection, that is robust to noise, is proposed. In the first phase, the hairline fractures are coarsely localized using statistical correlation and by exploiting the bilateral symmetry of the human mandible. In the second phase, the fractures are precisely identified and highlighted using a Markov Random Field (MRF) modeling approach coupled with Maximum A Posteriori probability (MAP) estimation. Gibbs sampling is used to maximize the posterior probability. Experimental results on Computer Tomography (CT) scans from real patients are presented.

KW - Computer tomography

KW - Gibbs sampling

KW - Hairline fractures

KW - MAP

KW - MRF

UR - http://www.scopus.com/inward/record.url?scp=34547215217&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547215217&partnerID=8YFLogxK

U2 - 10.1109/WACV.2007.28

DO - 10.1109/WACV.2007.28

M3 - Conference contribution

SN - 0769527949

SN - 9780769527949

T3 - Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007

BT - Proceedings - IEEE Workshop on Applications of Computer Vision, WACV 2007

ER -