Machine recognition and representation of neonatal facial displays of acute pain

  • Authors:

  • Sheryl Brahnam;Chao-Fa Chuang;Frank Y. Shih;Melinda R. Slack

  • Affiliations:

  • Department of Computer Information Systems, Missouri State University, 3rd Floor Glass Hall, 901 South National, Springfield, MO 65804, USA;Computer Vision Laboratory, College of Computing Sciences, New Jersey Institute of Technology, University Heights, Newark, NJ 07102, USA;Computer Vision Laboratory, College of Computing Sciences, New Jersey Institute of Technology, University Heights, Newark, NJ 07102, USA;Medical Director of Neonatology, St. John's Hospital, 1235 E. Cherokee, Springfield, MO 65804, USA

  • Venue:
  • Artificial Intelligence in Medicine
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Objective: It has been reported in medical literature that health care professionals have difficulty distinguishing a newborn's facial expressions of pain from facial reactions to other stimuli. Although a number of pain instruments have been developed to assist health professionals, studies demonstrate that health professionals are not entirely impartial in their assessment of pain and fail to capitalize on all the information exhibited in a newborn's facial displays. This study tackles these problems by applying three different state-of-the-art face classification techniques to the task of distinguishing a newborn's facial expressions of pain. Methods: The facial expressions of 26 neonates between the ages of 18h and 3 days old were photographed experiencing the pain of a heel lance and a variety of stressors, including transport from one crib to another (a disturbance that can provoke crying that is not in response to pain), an air stimulus on the nose, and friction on the external lateral surface of the heel. Three face classification techniques, principal component analysis (PCA), linear discriminant analysis (LDA), and support vector machine (SVM), were used to classify the faces. Results: In our experiments, the best recognition rates of pain versus nonpain (88.00%), pain versus rest (94.62%), pain versus cry (80.00%), pain versus air puff (83.33%), and pain versus friction (93.00%) were obtained from an SVM with a polynomial kernel of degree 3. The SVM outperformed two commonly used methods in face classification: PCA and LDA, each using the L"1 distance metric. Conclusion: The results of this study indicate that the application of face classification techniques in pain assessment and management is a promising area of investigation. on.