IEEE Trans Med Imaging. 2022 Sep;41(9):2304-2317. doi: 10.1109/TMI.2022.3161739. Epub 2022 Aug 31.
Most deep learning models for temporal regression directly output the estimation based on single input images, ignoring the relationships between different images. In this paper, we propose deep relation learning for regression, aiming to learn different relations between a pair of input images. Four non-linear relations are considered: "cumulative relation," "relative relation," "maximal relation" and "minimal relation." These four relations are learned simultaneously from one deep neural network which has two parts: feature extraction and relation regression. We use an efficient convolutional neural network to extract deep features from the pair of input images and apply a Transformer for relation learning. The proposed method is evaluated on a merged dataset with 6,049 subjects with ages of 0-97 years using 5-fold cross-validation for the task of brain age estimation. The experimental results have shown that the proposed method achieved a mean absolute error (MAE) of 2.38 years, which is lower than the MAEs of 8 other state-of-the-art algorithms with statistical significance (p<0.05) in paired T-test (two-side).
大多数用于时间回归的深度学习模型直接基于单张输入图像输出估计值，而忽略了不同图像之间的关系。在本文中，我们提出了回归的深度关系学习，旨在学习一对输入图像之间的不同关系。考虑了四种非线性关系：“累积关系”、“相对关系”、“最大关系”和“最小关系”。这四种关系由一个具有两部分的深度神经网络同时学习：特征提取和关系回归。我们使用高效的卷积神经网络从输入图像对中提取深度特征，并应用 Transformer 进行关系学习。该方法在一个包含 6049 名年龄在 0 到 97 岁的受试者的合并数据集上进行了评估，使用 5 折交叉验证进行脑龄估计任务。实验结果表明，该方法的平均绝对误差（MAE）为 2.38 岁，低于其他 8 种具有统计学意义的最先进算法的 MAE（p<0.05），在配对 T 检验（双侧）中。
