Skip to main content
QUICK REVIEW

[论文解读] Identifying Autism Spectrum Disorder Based on Individual-Aware Down-Sampling and Multi-Modal Learning

Li Pan, Jundong Liu|ArXiv.org|Sep 19, 2021
Autism Spectrum Disorder Research参考文献 52被引用 8
一句话总结

该论文提出了一种基于图的深度学习框架,通过整合个体感知的下采样与多模态学习,以提升基于fMRI数据的ASD诊断效果。通过将脑活动建模为分层图结构,并应用图池化与图卷积网络(GCNs)来捕捉功能连接性与区域活动,同时考虑表型异质性,该方法在ABIDE-I数据集上实现了87.62%的准确率与0.92的AUC,优于现有最先进方法。

ABSTRACT

Autism Spectrum Disorder(ASD) is a set of neurodevelopmental conditions that affect patients' social abilities. In recent years, many studies have employed deep learning to diagnose this brain dysfunction through functional MRI (fMRI). However, existing approaches solely focused on the abnormal brain functional connections but ignored the impact of regional activities. Due to this biased prior knowledge, previous diagnosis models suffered from inter-site measurement heterogeneity and inter-individual phenotypic differences. To address this issue, we propose a novel feature extraction method for fMRI that can learn a personalized lower-resolution representation of the entire brain networking regarding both the functional connections and regional activities. Specifically, we abstract the brain imaging as a graph structure and straightforwardly downsample it to substructures by hierarchical graph pooling. To further recalibrate the distribution of the extracted features under phenotypic information, we subsequently embed the sparse feature vectors into a population graph, where the hidden inter-subject heterogeneity and homogeneity are explicitly expressed as inter- and intra-community connectivity differences, and utilize Graph Convolutional Networks to learn the node embeddings. By these means, our framework can extract features directly and efficiently from the entire fMRI and be aware of implicit inter-individual variance. We have evaluated our framework on the ABIDE-I dataset with 10-fold cross-validation. The present model has achieved a mean classification accuracy of 87.62\% and a mean AUC of 0.92, better than the state-of-the-art methods.

研究动机与目标

  • 解决现有基于fMRI的ASD诊断方法仅关注功能连接性而忽略区域脑活动的局限性。
  • 降低fMRI基ASD分类中的跨站点测量异质性与跨个体表型变异性。
  • 开发一种统一的端到端框架,同时以个性化图结构表示建模功能连接性与区域活动。
  • 将非影像表型数据(如性别、年龄、站点)整合到模型中,以提升鲁棒性与泛化能力。
  • 在ABIDE-I fMRI数据集上展示优于现有最先进方法的性能表现。

提出的方法

  • 将大脑建模为图结构,其中节点代表脑区,边代表从fMRI时间序列数据中提取的功能连接。
  • 采用基于自注意力的图池化机制实现分层下采样,生成保留连接性与区域活动的个性化、低分辨率脑子图。
  • 将池化后的子图嵌入至群体图中,群体图中的群内与群间连接模式反映了群内与群间异质性。
  • 应用图卷积网络(GCNs)学习节点嵌入,利用表型信息重新校准特征分布。
  • 使用多层感知机(MLP)基于GCN学习的节点嵌入对受试者进行ASD或对照分类。
  • 在ABIDE-I数据集上采用10折交叉验证进行端到端训练。

实验结果

研究问题

  • RQ1能否通过联合建模功能连接性与区域脑活动的图基框架提升ASD诊断准确率?
  • RQ2通过图池化实现的个体感知下采样在多大程度上可减轻fMRI数据中个体间变异性与跨站点异质性的影响?
  • RQ3图卷积网络在多大程度上可整合表型信息(如性别、年龄、站点)以正则化特征学习并提升模型鲁棒性?
  • RQ4所提出方法在ABIDE-I fMRI数据集上的分类准确率与AUC是否优于现有最先进方法?
  • RQ5群体图中学习到的节点嵌入是否能揭示与性别或年龄等表型属性相对应的有意义聚类模式?

主要发现

  • 所提框架在ABIDE-I数据集上实现了87.62% ± 0.06的平均分类准确率与0.92 ± 0.01的平均AUC,优于现有最先进方法。
  • 图卷积网络(GCN)分类器表现出更优的鲁棒性,其准确率(0.027)与AUC(0.0037)的标准差更低,优于逻辑回归。
  • GCN性能优于逻辑回归(86.07%准确率与0.93 AUC),但训练时间显著更长(76,169个周期 vs. 625个周期)。
  • GCN学习的节点嵌入经t-SNE可视化后,基于表型属性(尤其是性别)显示出清晰的聚类模式,表明有效捕捉了个体间差异。
  • 模型在各折中表现稳定,准确率与AUC方差较低,表明具备强泛化能力与鲁棒性。
  • 将表型信息整合至群体图中,改善了特征重校准,增强了模型泛化能力,尤其在处理跨站点变异性方面表现突出。

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。