Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future

Abstract

With the advances of data-driven machine learning research, a wide variety of prediction problems have been tackled. It has become critical to explore how machine learning and specifically deep learning methods can be exploited to analyse healthcare data. A major limitation of existing methods has been the focus on grid-like data; however, the structure of physiological recordings are often irregular and unordered which makes it difficult to conceptualise them as a matrix. As such, graph neural networks have attracted significant attention by exploiting implicit information that resides in a biological system, with interactive nodes connected by edges whose weights can be either temporal associations or anatomical junctions. In this survey, we thoroughly review the different types of graph architectures and their applications in healthcare. We provide an overview of these methods in a systematic manner, organized by their domain of application including functional connectivity, anatomical structure and electrical-based analysis. We also outline the limitations of existing techniques and discuss potential directions for future research.

References in corpus (42)

• Semi-Supervised Classification with Graph Convolutional Networks
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Striving for Simplicity: The All Convolutional Net
• Simplifying Graph Convolutional Networks
• REFUGE Challenge: A Unified Framework for Evaluating Automated Methods for Glaucoma Assessment from Fundus Photographs
• A large annotated medical image dataset for the development and evaluation of segmentation algorithms
• Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
• FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling
• Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning
• Contrastive Multi-View Representation Learning on Graphs
• Self-Attention Graph Pooling
• Standardized Assessment of Automatic Segmentation of White Matter Hyperintensities and Results of the WMH Segmentation Challenge
• GraphGAN: Graph Representation Learning with Generative Adversarial Nets
• GraphVite: A High-Performance CPU-GPU Hybrid System for Node Embedding
• Relation Modeling with Graph Convolutional Networks for Facial Action Unit Detection
• Graph WaveNet for Deep Spatial-Temporal Graph Modeling
• TADPOLE Challenge: Accurate Alzheimer's disease prediction through crowdsourced forecasting of future data
• Spherical CNNs on Unstructured Grids
• When Does Self-Supervision Help Graph Convolutional Networks?
• Human Pose Estimation with Spatial Contextual Information
• Multi-Label Image Recognition with Graph Convolutional Networks
• Explainability Techniques for Graph Convolutional Networks
• Uncertainty Aware Semi-Supervised Learning on Graph Data
• Temporal Graph Convolutional Networks for Automatic Seizure Detection
• EEG-GCNN: Augmenting Electroencephalogram-based Neurological Disease Diagnosis using a Domain-guided Graph Convolutional Neural Network
• GraphFL: A Federated Learning Framework for Semi-Supervised Node Classification on Graphs
• Towards Explainable Graph Representations in Digital Pathology
• Classification of Medical Images and Illustrations in the Biomedical Literature Using Synergic Deep Learning
• Fast Interactive Object Annotation with Curve-GCN
• Attention-based Graph ResNet for Motor Intent Detection from Raw EEG signals
• Zoom in to where it matters: a hierarchical graph based model for mammogram analysis
• DeepSphere: a graph-based spherical CNN
• Temporal Graph Modeling for Skeleton-based Action Recognition
• Differentiable Hierarchical Graph Grouping for Multi-Person Pose Estimation
• Uncertainty quantification of molecular property prediction with Bayesian neural networks
• Graph Neural Network for Interpreting Task-fMRI Biomarkers
• GOSH: Embedding Big Graphs on Small Hardware
• MER-GCN: Micro Expression Recognition Based on Relation Modeling with Graph Convolutional Network
• Deep neural networks on graph signals for brain imaging analysis
• Lymph Node Gross Tumor Volume Detection in Oncology Imaging via Relationship Learning Using Graph Neural Network
• An Uncertainty-Driven GCN Refinement Strategy for Organ Segmentation
• Learnable Pooling in Graph Convolution Networks for Brain Surface Analysis