Subclass-based multi-task learning for Alzheimer's disease diagnosis

Heung-II Suk, Seong-Whan Lee, Dinggang Shen, The Alzheimers Disease Neuroimaging Initiative

In this work, we propose a novel subclass-based multi-task learning method for feature selection in computer-aided Alzheimer's Disease (AD) or Mild Cognitive Impairment (MCI) diagnosis. Unlike the previous methods that often assumed a unimodal data distribution, we take into account the underlying multipeak distribution of classes. The rationale for our approach is that it is highly likely for neuroimaging data to have multiple peaks or modes in distribution, e.g., mixture of Gaussians, due to the inter-subject variability. In this regard, we use a clustering method to discover the multipeak distributional characteristics and define subclasses based on the clustering results, in which each cluster covers a peak in the underlying multipeak distribution. Specifically, after performing clustering for each class, we encode the respective subclasses, i.e., clusters, with their unique codes. In encoding, we impose the subclasses of the same original class close to each other and those of different original classes distinct from each other. By setting the codes as new label vectors of our training samples, we formulate a multi-task learning problem in a ℓ2,1-penalized regression framework, through which we finally select features for classification. In our experimental results on the ADNI dataset, we validated the effectiveness of the proposed method by improving the classification accuracies by 1% (AD vs. Normal Control: NC), 3.25% (MCI vs. NC), 5.34% (AD vs. MCI), and 7.4% (MCI Converter: MCI-C vs. MCI Non-Converter: MCI-NC) compared to the competing single-task learning method. It is remarkable for the performance improvement in MCI-C vs. MCI-NC classification, which is the most important for early diagnosis and treatment. It is also noteworthy that with the strategy of modality-adaptive weights by means of a multi-kernel support vector machine, we maximally achieved the classification accuracies of 96.18% (AD vs. NC), 81.45% (MCI vs. NC), 73.21% (AD vs. MCI), and 74.04% (MCI-C vs. MCI-NC), respectively.