A neural network model for early diagnosis of acute GVHD based on gene expression data

Acute graft-versus-host disease (aGVHD) is the major complication after allogeneic haematopoietic stem cell transplantation (HSCT) in which functional immune cells of donor recognize the recipient as ldquoforeignrdquo and mount an immunologic attack. In this paper we analyzed gene-expression profiles of 47 genes associated with alloreactivity in 59 patients submitted to HSCT. We have applied a dimension reduction technique to found the most important subset of genes to make a diagnosis of aGVHD. The composed subset has been used in order to train and test a suitable artificial neural network (ANN) to detect the aGVHD at on-set of clinical signs.