Automatic facial analysis for objective assessment of facial paralysis

J.J. Soraghan, Brian O'Reilly, Shu He, Stewart McGrenary

Research output: Contribution to conferenceKeynote

Abstract

Facial Paralysis is a condition causing decreased movement on one side of the face. A quantitative, objective and reliable assessment system would be an invaluable tool for clinicians treating patients with this condition. This paper presents an approach based on the automatic analysis of patient video data. Facial feature localization and facial movement detection methods are discussed. An algorithm is presented to process the optical flow data to obtain the motion features in the relevant facial regions. Three classification methods are applied to provide quantitative evaluations of regional facial nerve function and the overall facial nerve function based on the House-Brackmann Scale. Experiments show the Radial Basis Function (RBF) Neural Network to have superior performance.

Conference

Conference1st International Conferfence on Computer Science from Algorithms to Applications (CSAA-2009)
CityCairo, Egypt
Period8/01/0910/01/09

Fingerprint

Optical flows
Neural networks
Experiments

Keywords

  • facial paralysis measurement
  • house-brackmann scale
  • optical flow
  • rbf neural network

Cite this

Soraghan, J. J., O'Reilly, B., He, S., & McGrenary, S. (2009). Automatic facial analysis for objective assessment of facial paralysis. 1st International Conferfence on Computer Science from Algorithms to Applications (CSAA-2009), Cairo, Egypt, .
Soraghan, J.J. ; O'Reilly, Brian ; He, Shu ; McGrenary, Stewart. / Automatic facial analysis for objective assessment of facial paralysis. 1st International Conferfence on Computer Science from Algorithms to Applications (CSAA-2009), Cairo, Egypt, .
@conference{d7f96e92268d40ab929c7ef14f8036e2,
title = "Automatic facial analysis for objective assessment of facial paralysis",
abstract = "Facial Paralysis is a condition causing decreased movement on one side of the face. A quantitative, objective and reliable assessment system would be an invaluable tool for clinicians treating patients with this condition. This paper presents an approach based on the automatic analysis of patient video data. Facial feature localization and facial movement detection methods are discussed. An algorithm is presented to process the optical flow data to obtain the motion features in the relevant facial regions. Three classification methods are applied to provide quantitative evaluations of regional facial nerve function and the overall facial nerve function based on the House-Brackmann Scale. Experiments show the Radial Basis Function (RBF) Neural Network to have superior performance.",
keywords = "facial paralysis measurement, house-brackmann scale, optical flow, rbf neural network",
author = "J.J. Soraghan and Brian O'Reilly and Shu He and Stewart McGrenary",
year = "2009",
month = "12",
language = "English",
note = "1st International Conferfence on Computer Science from Algorithms to Applications (CSAA-2009) ; Conference date: 08-01-2009 Through 10-01-2009",

}

Soraghan, JJ, O'Reilly, B, He, S & McGrenary, S 2009, 'Automatic facial analysis for objective assessment of facial paralysis' 1st International Conferfence on Computer Science from Algorithms to Applications (CSAA-2009), Cairo, Egypt, 8/01/09 - 10/01/09, .

Automatic facial analysis for objective assessment of facial paralysis. / Soraghan, J.J.; O'Reilly, Brian; He, Shu; McGrenary, Stewart.

2009. 1st International Conferfence on Computer Science from Algorithms to Applications (CSAA-2009), Cairo, Egypt, .

Research output: Contribution to conferenceKeynote

TY - CONF

T1 - Automatic facial analysis for objective assessment of facial paralysis

AU - Soraghan, J.J.

AU - O'Reilly, Brian

AU - He, Shu

AU - McGrenary, Stewart

PY - 2009/12

Y1 - 2009/12

N2 - Facial Paralysis is a condition causing decreased movement on one side of the face. A quantitative, objective and reliable assessment system would be an invaluable tool for clinicians treating patients with this condition. This paper presents an approach based on the automatic analysis of patient video data. Facial feature localization and facial movement detection methods are discussed. An algorithm is presented to process the optical flow data to obtain the motion features in the relevant facial regions. Three classification methods are applied to provide quantitative evaluations of regional facial nerve function and the overall facial nerve function based on the House-Brackmann Scale. Experiments show the Radial Basis Function (RBF) Neural Network to have superior performance.

AB - Facial Paralysis is a condition causing decreased movement on one side of the face. A quantitative, objective and reliable assessment system would be an invaluable tool for clinicians treating patients with this condition. This paper presents an approach based on the automatic analysis of patient video data. Facial feature localization and facial movement detection methods are discussed. An algorithm is presented to process the optical flow data to obtain the motion features in the relevant facial regions. Three classification methods are applied to provide quantitative evaluations of regional facial nerve function and the overall facial nerve function based on the House-Brackmann Scale. Experiments show the Radial Basis Function (RBF) Neural Network to have superior performance.

KW - facial paralysis measurement

KW - house-brackmann scale

KW - optical flow

KW - rbf neural network

UR - http://www.csaa09.com/

M3 - Keynote

ER -

Soraghan JJ, O'Reilly B, He S, McGrenary S. Automatic facial analysis for objective assessment of facial paralysis. 2009. 1st International Conferfence on Computer Science from Algorithms to Applications (CSAA-2009), Cairo, Egypt, .