Predicting the age of healthy adults from structural MRI by sparse representation

  • Authors:

  • Longfei Su;Lubin Wang;Dewen Hu

  • Affiliations:

  • College of Mechatronics and Automation, National University of Defense Technology, Changsha, Hunan, China;College of Mechatronics and Automation, National University of Defense Technology, Changsha, Hunan, China;College of Mechatronics and Automation, National University of Defense Technology, Changsha, Hunan, China

  • Venue:
  • IScIDE'12 Proceedings of the third Sino-foreign-interchange conference on Intelligent Science and Intelligent Data Engineering
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

It is generally accepted that degenerative brain diseases lead to abnormal aging process of the human brain. Thus, healthy brain aging model has great potential in clinical diagnosis and intervention. The aim of this work is to construct a regression model which is efficient for age prediction of healthy brain. Two groups of T1-weighted MRI images were involved. The first group was used for voxel selection then corresponding voxels in the second group were used for age prediction. Then mean absolute error (MAE) between the predicted age and the true age is obtained. The age prediction accuracy can reach as high as 4.67 years (MAE). In conclusion, the framework in current study can be a healthy aging model for abnormality detection of human brain. The brain regions identified by this model is sensitive to aging process which can be viewed as biomarker of brain age.