Path Signature Neural Network of Cortical Features for Prediction of Infant Cognitive Scores.
Path Signature Neural Network of Cortical Features for Prediction of Infant Cognitive Scores.
复制标题
用于预测婴儿认知评分的皮质特征的路径签名神经网络。
DOI：
10.1109/tmi.2022.3147690
复制
发表时间：
2022-07
影响因子：
10.6
通讯作者：
中科院分区：
文献类型：
--
作者：
研究表明，婴儿期的认知技能和大脑形态之间存在密切联系。然而，考虑到特征维数过大、样本量小和数据缺失等问题，利用个体的脑形态特征预测个体的认知成绩仍然是一个巨大的挑战。由于有限的数据，一个紧凑但富有表现力的特征集是可取的，因为它可以降低维数，避免潜在的过拟合问题。因此，我们开创了路径签名方法，以进一步探索纵向皮质特征的基本隐藏的动态模式。为了形成一个层次化的和更丰富的时间表示，在这项工作中，提出了一种新的基于皮层特征的路径签名神经网络（CF-PSNet）与堆叠的可区分的时间路径签名层预测的个人认知分数。通过在路径生成中引入存在嵌入，可以提高算法对数据丢失的鲁棒性。得益于CF-PSNet的全局时间感受野，可以充分利用现有数据中包含的特性。此外，由于不需要整个大脑为特定的认知能力而工作，因此使用前K选择模块来选择最有影响力的大脑区域，从而降低模型大小和过拟合的风险。在9个时间点内对内部纵向婴儿数据集进行了广泛的实验。通过与最近的几种算法进行比较，我们说明了我们的CF-PSNet的最先进的性能（即，均方根误差为0.027，每个样本的时间延迟为518毫秒）。
Studies have shown that there is a tight connection between cognition skills and brain morphology during infancy. Nonetheless, it is still a great challenge to predict individual cognitive scores using their brain morphological features, considering issues like the excessive feature dimension, small sample size and missing data. Due to the limited data, a compact but expressive feature set is desirable as it can reduce the dimension and avoid the potential overfitting issue. Therefore, we pioneer the path signature method to further explore the essential hidden dynamic patterns of longitudinal cortical features. To form a hierarchical and more informative temporal representation, in this work, a novel cortical feature based path signature neural network (CF-PSNet) is proposed with stacked differentiable temporal path signature layers for prediction of individual cognitive scores. By introducing the existence embedding in path generation, we can improve the robustness against the missing data. Benefiting from the global temporal receptive field of CF-PSNet, characteristics consisted in the existing data can be fully leveraged. Further, as there is no need for the whole brain to work for a certain cognitive ability, a top K selection module is used to select the most influential brain regions, decreasing the model size and the risk of overfitting. Extensive experiments are conducted on an in-house longitudinal infant dataset within 9 time points. By comparing with several recent algorithms, we illustrate the state-of-the-art performance of our CF-PSNet (i.e., root mean square error of 0.027 with the time latency of 518 milliseconds for each sample).
DOI：
10.1007/978-3-030-59728-3_14
发表时间：
2020-10
期刊：
Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
影响因子：
--
作者：
Zhang X;Cheng J;Ni H;Li C;Xu X;Wu Z;Wang L;Lin W;Shen D;Li G
通讯作者：
Li G