Liu Mingxia, Zhang Jun, Adeli Ehsan, Shen Dinggang
Department of Radiology and BRIC, University of North Carolina at Chapel Hill, North Carolina 27599, U.S.A.
Med Image Comput Comput Assist Interv. 2017 Sep;10435:3-11. doi: 10.1007/978-3-319-66179-7_1. Epub 2017 Sep 4.
Jointly identifying brain diseases and predicting clinical scores have attracted increasing attention in the domain of computer-aided diagnosis using magnetic resonance imaging (MRI) data, since these two tasks are highly correlated. Although several joint learning models have been developed, most existing methods focus on using human-engineered features extracted from MRI data. Due to the possible heterogeneous property between human-engineered features and subsequent classification/regression models, those methods may lead to sub-optimal learning performance. In this paper, we propose a deep multi-task multi-channel learning (DML) framework for simultaneous classification and regression for brain disease diagnosis, using MRI data and personal information (, age, gender, and education level) of subjects. Specifically, we first identify discriminative anatomical landmarks from MR images in a data-driven manner, and then extract multiple image patches around these detected landmarks. A deep multi-task multi-channel convolutional neural network is then developed for joint disease classification and clinical score regression. We train our model on a large multi-center cohort (, ADNI-1) and test it on an cohort (, ADNI-2). Experimental results demonstrate that DML is superior to the state-of-the-art approaches in brain diasease diagnosis.
由于脑疾病识别和临床评分预测这两项任务高度相关，因此在利用磁共振成像（MRI）数据的计算机辅助诊断领域，它们正吸引着越来越多的关注。尽管已经开发了几种联合学习模型，但大多数现有方法都侧重于使用从MRI数据中提取的人工特征。由于人工特征与后续分类/回归模型之间可能存在异质性，这些方法可能会导致次优的学习性能。在本文中，我们提出了一种深度多任务多通道学习（DML）框架，用于利用MRI数据和受试者的个人信息（年龄、性别和教育水平）对脑疾病诊断进行同步分类和回归。具体而言，我们首先以数据驱动的方式从MR图像中识别出有鉴别力的解剖标志，然后在这些检测到的标志周围提取多个图像块。接着，我们开发了一个深度多任务多通道卷积神经网络，用于联合疾病分类和临床评分回归。我们在一个大型多中心队列（ADNI-1）上训练我们的模型，并在另一个队列（ADNI-2）上进行测试。实验结果表明，DML在脑疾病诊断方面优于现有最先进的方法。
