A Comparative Study of Population-Graph Construction Methods and Graph Neural Networks for Brain Age Regression

被引:0
|
作者
Bintsi, Kyriaki-Margarita [1 ]
Mueller, Tamara T. [2 ]
Starck, Sophie [2 ]
Baltatzis, Vasileios [1 ,3 ]
Hammers, Alexander [3 ]
Rueckert, Daniel [1 ,2 ]
机构
[1] Imperial Coll London, Dept Comp, BioMedIA, London, England
[2] Tech Univ Munich, Fac Informat, Lab AI Med & Healthcare, Munich, Germany
[3] Kings Coll London, Biomed Engn & Imaging Sci, London, England
基金
英国工程与自然科学研究理事会;
关键词
Brain age regression; Population graphs; Graph Neural Networks;
D O I
10.1007/978-3-031-55088-1_6
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The difference between the chronological and biological brain age of a subject can be an important biomarker for neurodegenerative diseases, thus brain age estimation can be crucial in clinical settings. One way to incorporate multimodal information into this estimation is through population graphs, which combine various types of imaging data and capture the associations among individuals within a population. In medical imaging, population graphs have demonstrated promising results, mostly for classification tasks. In most cases, the graph structure is pre-defined and remains static during training. However, extracting population graphs is a non-trivial task and can significantly impact the performance of Graph Neural Networks (GNNs), which are sensitive to the graph structure. In this work, we highlight the importance of a meaningful graph construction and experiment with different population-graph construction methods and their effect on GNN performance on brain age estimation. We use the homophily metric and graph visualizations to gain valuable quantitative and qualitative insights on the extracted graph structures. For the experimental evaluation, we leverage the UK Biobank dataset, which offers many imaging and non-imaging phenotypes. Our results indicate that architectures highly sensitive to the graph structure, such as Graph Convolutional Network (GCN) and Graph Attention Network (GAT), struggle with low homophily graphs, while other architectures, such as GraphSage and Chebyshev, are more robust across different homophily ratios. We conclude that static graph construction approaches are potentially insufficient for the task of brain age estimation and make recommendations for alternative research directions.
引用
收藏
页码:64 / 73
页数:10
相关论文
共 50 条
  • [1] Multimodal Brain Age Estimation Using Interpretable Adaptive Population-Graph Learning
    Bintsi, Kyriaki-Margarita
    Baltatzis, Vasileios
    Potamias, Rolandos Alexandros
    Hammers, Alexander
    Rueckert, Daniel
    MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION, MICCAI 2023, PT VIII, 2023, 14227 : 195 - 204
  • [2] Graph Neural Networks for Brain Graph Learning: A Survey
    Luo, Xuexiong
    Wu, Jia
    Yang, Jian
    Xue, Shan
    Beheshti, Amin
    Sheng, Quan Z.
    McAlpine, David
    Sowman, Paul
    Giral, Alexis
    Yu, Philip S.
    PROCEEDINGS OF THE THIRTY-THIRD INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2024, 2024, : 8170 - 8178
  • [3] Graph neural networks for construction applications
    Jia, Yilong
    Wang, Jun
    Shou, Wenchi
    Hosseini, M. Reza
    Bai, Yu
    AUTOMATION IN CONSTRUCTION, 2023, 154
  • [4] Explaining the Explainers in Graph Neural Networks: a Comparative Study
    Longa, Antonio
    Azzolin, Steve
    Santin, Gabriele
    Cencetti, Giulia
    Lio, Pietro
    Lepri, Bruno
    Passerini, Andrea
    ACM COMPUTING SURVEYS, 2025, 57 (05)
  • [5] A Comparative Study of Graph Neural Networks for Shape Classification in Neuroimaging
    Shehata, Nairouz
    Bain, Wulfie
    Glocker, Ben
    GEOMETRIC DEEP LEARNING IN MEDICAL IMAGE ANALYSIS, VOL 194, 2022, 194 : 160 - 171
  • [6] Graph pooling in graph neural networks: methods and their applications in omics studies
    Wang, Yan
    Hou, Wenju
    Sheng, Nan
    Zhao, Ziqi
    Liu, Jialin
    Huang, Lan
    Wang, Juexin
    ARTIFICIAL INTELLIGENCE REVIEW, 2024, 57 (11)
  • [7] On Graph Construction for Classification of Clinical Trials Protocols Using Graph Neural Networks
    Ferdowsi, Sohrab
    Copara, Jenny
    Gouareb, Racha
    Borissov, Nikolay
    Jaume-Santero, Fernando
    Amini, Poorya
    Teodoro, Douglas
    ARTIFICIAL INTELLIGENCE IN MEDICINE, AIME 2022, 2022, 13263 : 249 - 259
  • [8] A Demonstration of Interpretability Methods for Graph Neural Networks
    Mobaraki, Ehsan B.
    Khan, Arijit
    PROCEEDINGS OF THE 6TH ACM SIGMOD JOINT INTERNATIONAL WORKSHOP ON GRAPH DATA MANAGEMENT EXPERIENCES & SYSTEMS AND NETWORK DATA ANALYTICS, GRADES-NDA 2023, 2023,
  • [9] Graph neural networks: A review of methods and applications
    Zhou, Jie
    Cui, Ganqu
    Hu, Shengding
    Zhang, Zhengyan
    Yang, Cheng
    Liu, Zhiyuan
    Wang, Lifeng
    Li, Changcheng
    Sun, Maosong
    AI OPEN, 2020, 1 : 57 - 81
  • [10] Graph neural networks: A review of methods and applications
    Zhou, Jie
    Cui, Ganqu
    Hu, Shengding
    Zhang, Zhengyan
    Yang, Cheng
    Liu, Zhiyuan
    Wang, Lifeng
    Li, Changcheng
    Sun, Maosong
    AI OPEN, 2020, 1 : 57 - 81