论文全名:Multi-Scale FC-Based Multi-Order GCN: A Novel Model for Predicting Individual Behavior From fMRI
英文是纯手打的!论文原文的summarizing and paraphrasing。可能会出现难以避免的拼写错误和语法错误,若有发现欢迎评论指正!文章偏向于笔记,谨慎食用
目录
2.4. Multi-Scale Brain-Behavior Relationship
2.5. Multi-Scale FC Based Multi-Order Graph Convolutional Network
2.5.1. Multi-Scale Functional Connectivity Estimation
2.5.2. Multi-Order Graph Convolutional Network
2.5.3. Adaptive Feature Fusion
2.5.4. Behavior Score Estimation
2.6.1. Model Settings and Evaluation Metric
2.7.1. Comparison of FC-Behavior Relationship Between Different Scales
2.7.3. Comparison With Other Methods
2.7.5. Importance of Functional Connectivity
1. 心得
(1)公式中notation的上下标多得令人不适...虽然没什么问题,而且也易懂,就是看上去太大一坨了哈哈哈哈哈
(2)Trainable parameter实在是有点多hhhh,可能会跑得久一点的亚子
(3)怎么老是定义函数啊,直接写出来名字或者数学表达也不是不行
(4)虽然没有代码,但是是较为容易复现的论文
2. 论文逐段精读
2.1. Abstract
①They proposed a Multi-Scale FC-based Multi-Order GCN (MSFC-MO-GCN)
②⭐There are few researches on behaviour prediction
2.2. Introduction
①作者认为“GCN仅根据单尺度(即单空间分辨率)FCN 学习大脑连接表示”。持否定意见,也有使用GCN+超图的或GCN+不同脑图谱,而且有时候其他尺度来源于模态,受到了数据集本身的限制
②作者认为“图卷积层只利用了来自节点的 1 阶邻居的潜在信息,忽略了脑区之间远距离功能交互产生的丰富信息”。持否定意见,这取决于脑网络如何构建。一个全连接的脑网络不存在“近端脑区”或“远端脑区”
③作者认为“例如,在具有粗粒度类的分类任务中,高阶近似可能比低阶近似更有用”。感觉高阶近似大概率都比低阶近似好吧?大家不都是为了简化模型流程以及降低运算时间吗?
④Thus, they proposed a Multi-Scale FC based Multi-Order GCN (MSFC-MO-GCN)
lattice n. 格子木架,格子金属架,格栅(用作篱笆等);斜条结构;斜格图案
granularity n. 粒度;(颗,成)粒性
2.3. Materials
2.3.1. Data Information
①Dataset: HCP S1200
②Screen criterion: no neuropsychiatric disorders
③Experimental design: staring the bright cross-hair on a dark background with relaxed state and eyes open
④Samples: 805
⑤5 tasks chosen: one motor-related test (Endurance), one executive-function-related test (Cognitive Flexibility), one memory-related test (Episodic Memory), one language-related test (Story Difficulty Level), and a comprehensive cognitive test (Fluid Intelligence)
⑥Measurement: NIH Cognition Battery toolbox
⑦Score adjustment: using NIH National Norms toolbox to adjust all the scores to a standard deviation
2.3.2. Imaging Preprocessing
①Minimal preprocessing pipeline: HCP fMRIvolume
②Pre-processing process: 1) gradient distortion correction, 2) head motion correction, 3) EPI distortion correction, 4) registration to the Montreal Neurological Institute (MNI) space, 5) intensity normalization to a global mean, and 6) masking out non-brain voxels
③Artifact removal: independent component analysis (ICA) based FIX Xnoiseifier
2.4. Multi-Scale Brain-Behavior Relationship
①Atlas: Schaefer 100, 500 and 1k
②FCN construction: Pearson correlation
③Connections: all the negative connections are set to 0 and only remains the top 5% high value
④System-level analysis: 7 functional subsystems, visual network (VIS), somatosensory-motor network (SM), attention network (ATT), salience network (SAL), limbic system (LIM), frontoparietal network (FP), and default mode network (DMN), with left and right brain, obtaining 14 regions overall.
⑤Calculating the Pearson correlation between each network and applying FDR correction
⑥CS matrix: The diagonal of the matrix represents the CS between systems and behaviors, while non diagonal values represent the FC between systems and CS between behaviors
2.5. Multi-Scale FC Based Multi-Order Graph Convolutional Network
①Overall framework:
2.5.1. Multi-Scale Functional Connectivity Estimation
①Sparsify: only remain 5 strong edges for each node for ensuring the connectivity of the graph(为什么又说“仅保留值最高的5%边缘”,又说“每个节点保留五个最强边?“但不是总共就14个节点吗...)
②Scales of one graph:
2.5.2. Multi-Order Graph Convolutional Network
①Graph for the -th subject:
②Graph at the -th scale:
(1)Multi-Order Graph Convolution Layer
①Multi-order aggregation(这个只卷积了一次,是一个人的其中一个scale(atlas),三个颜色是指邻居阶数的不同。作者定义几阶的邻接矩阵就是把原始的邻接矩阵乘阶数的次幂,0阶的时候是单位矩阵):
②Node feature
③Graph convolution operator:
and the define as ReLU......非常独特的消息传递方式......
④
(2)Pooling Layer
①The final feature vector can be calculated by:
(3)Inter-Scale Contrast Constraint
①To enhance the similarity between different scales:
where the features from the same scale are positive term and from different scale are negative term, denotes Euclidean distance between two vectors, denotes margin parameter
2.5.3. Adaptive Feature Fusion
①Total feature of subject :
②Mean pooling to obtain graph representation:
where denotes global average pooling
③Contribution weight (attention?):
where and are trainable parameters, is Sigmoid function
④Joint/final features for one person:
2.5.4. Behavior Score Estimation
①Behavior score:
where denotes trainable parameter
②Loss:
where denotes absolute error
2.6. Implementation
2.6.1. Model Settings and Evaluation Metric
①Filters in 2 GCN: 96 and 12(GCN还有滤波器吗???这什么玩意儿?hidden layer?)
②12-channel pooling layer and a 1-channel fully connected layer
③Optimizer: Adam
④Learning rate: 0.005
⑤Norm: L2 with 0.0005
⑥Cross validation: 5 fold
⑦Batch size: 16
⑧Iteration times: 70
⑨Evaluation: average value in 5 times of 5-fold cross validation
2.6.2. Compared Methods
①Hyperparameter setting in each compared method:
(1)Kernel Regression Method
(2)FNN
(3)BrainNetCNN
(4)GCNN
(5)GAT
(6)SAGPool
(7)Meta-RegGNN
(8)BC-GCN-SE
2.7. Results and Discussion
2.7.1. Comparison of FC-Behavior Relationship Between Different Scales
①Correlation difference between scales:
so they reckon the brain has a hierarchical structure
2.7.2. Parameter Analysis
①Grid search on hyperparameter:
where denotes the order in each scale
②Fixing to 1 and further testing the combinations of :
2.7.3. Comparison With Other Methods
①Comparison table:
2.7.4. Ablation Study
①Module ablation:
where baseline 1 denotes Single-Scale FCs + Multi-Order Graph Convolution, 2 denotes Multi-Scale FCs + 1-Order Graph Convolution + Inter-Scale Contrast Constraint, and 3 represents Multi-Scale FCs + Multi-Order Graph Convolution
2.7.5. Importance of Functional Connectivity
①Applying occlusion importance (OI) in networks (屏蔽某一个网络的特征得到的结果和原始结果的差异):
2.8. Conclusion
~
3. 知识补充
3.1. FDR correction
(1)定义:FDR correction,即错误发现率(False Discovery Rate)校正,是一种在多重假设检验中常用的统计校正方法,旨在控制假阳性(false positives)发现的错误率。
(2)方法:FDR是指在拒绝原假设的条件下,拒绝的假设中错误的比例。FDR correction则是一种通过调整统计显著性水平来降低这一比例的方法。FDR是指在拒绝原假设的条件下,拒绝的假设中错误的比例。FDR correction则是一种通过调整统计显著性水平来降低这一比例的方法。
4. Reference
Wen, X. et al. (2024) 'Multi-Scale FC-Based Multi-Order GCN: A Novel Model for Predicting Individual Behavior From fMRI', IEEE Transactions on Neural Systems and Rehabilitation Engineering , 32: 548-558. doi: 10.1109/TNSRE.2024.3357059