Detecting Active Brain Regions by a Constrained Alternating Least Squares Nonnegative Matrix Factorization Algorithm from Single Subject's fMRI Data

단일 대상의 fMRI 데이터에서 제약적 교차 최소 제곱 비음수 행렬 분해 알고리즘에 의한 활성화 뇌 영역 검출

  • Published : 2011.06.29

Abstract

In this paper, we propose a constrained alternating least squares nonnegative matrix factorization algorithm (cALSNMF) to detect active brain regions from single subject's task-related fMRI data. In cALSNMF, we define a new cost function which considers the uncorrelation and noisy problems of fMRI data by adding decorrelation and smoothing constraints in original Euclidean distance cost function. We also generate a novel training procedure by modifying the update rules and combining with optimal brain surgeon (OBS) algorithm. The experimental results on visuomotor task fMRI data show that our cALSNMF fits fMRI data better than original ALSNMF in detecting task-related brain activation from single subject's fMRI data.

Keywords

Acknowledgement

Supported by : National Research Foundation of Korea (NRF)