Computer Science and Engineering, Northeastern University, Shenyang, China.
Computer Science and Engineering, Northeastern University, Shenyang, China.
Comput Methods Programs Biomed. 2018 Aug;162:19-45. doi: 10.1016/j.cmpb.2018.04.028. Epub 2018 May 3.
Alzheimers disease (AD) is characterized by gradual neurodegeneration and loss of brain function, especially for memory during early stages. Regression analysis has been widely applied to AD research to relate clinical and biomarker data such as predicting cognitive outcomes from Magnetic Resonance Imaging (MRI) measures. Recently, the multi-task feature learning (MTFL) methods have been widely studied to predict cognitive outcomes and select the discriminative feature subset from MRI features by incorporating inherent correlations among multiple clinical cognitive measures. However, the existing MTFL assumes the correlation among all the tasks is uniform, and the task relatedness is modeled by encouraging a common subset of features with neglecting the inherent structure of tasks and MRI features.
In this paper, we proposed a generalized fused group lasso (GFGL) regularization to model the underlying structures, involving (1) a graph structure within tasks and (2) a group structure among the image features. Then, we present a multi-task learning framework (called GFGL-MTFL), combining the ℓ-norm with the GFGL regularization, to model the flexible structures.
Through empirical evaluation and comparison with different baseline methods and the state-of-the-art MTL methods on data from Alzheimer's Disease Neuroimaging Initiative (ADNI) database, we illustrate that the proposed GFGL-MTFL method outperforms other methods in terms of both Mean Squared Error (nMSE) and weighted correlation coefficient (wR). Improvements are statistically significant for most scores (tasks).
The experimental results with real and synthetic data demonstrate that incorporating the two prior structures by the generalized fused group lasso norm into the multi task feature learning can improve the prediction performance over several state-of-the-art competing methods, and the estimated correlation of the cognitive functions and the identification of cognition relevant imaging markers are clinically and biologically meaningful.
阿尔茨海默病（AD）的特征是逐渐的神经退行性变和脑功能丧失，尤其是在早期的记忆。回归分析已广泛应用于 AD 研究，以将临床和生物标志物数据相关联，例如从磁共振成像（MRI）测量结果预测认知结果。最近，多任务特征学习（MTFL）方法已被广泛研究，以通过合并多个临床认知测量中的固有相关性，从 MRI 特征中预测认知结果并选择有区别的特征子集。但是，现有的 MTFL 假设所有任务之间的相关性是均匀的，并且通过鼓励具有共同特征子集的方式来对任务相关性进行建模，而忽略了任务和 MRI 特征的内在结构。
在本文中，我们提出了一种广义融合组套索（GFGL）正则化方法来建模潜在结构，包括（1）任务内的图结构和（2）图像特征之间的组结构。然后，我们提出了一种多任务学习框架（称为 GFGL-MTFL），将 ℓ-范数与 GFGL 正则化相结合，以对灵活的结构进行建模。
通过对来自阿尔茨海默氏病神经影像学倡议（ADNI）数据库的数据的实证评估和与不同基线方法以及最新的 MTL 方法的比较，我们说明了所提出的 GFGL-MTFL 方法在均方误差（nMSE）和加权相关系数（wR）方面均优于其他方法。对于大多数分数（任务），改进具有统计学意义。
真实和合成数据的实验结果表明，通过广义融合组套索范数将两种先验结构纳入多任务特征学习中，可以提高预测性能，超过几个最新的竞争方法，并且认知功能的估计相关性和识别认知相关的影像学标志物在临床上和生物学上均具有意义。
