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

Ahmedt-Aristizabal, David; Armin, Mohammad Ali; Denman, Simon; Fookes, Clinton; Petersson, Lars

doi:10.48550/arxiv.2105.13137

Cited by 4 publications

(6 citation statements)

References 249 publications

(452 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Beyond generating predictions relating to biology and medicine at molecular, genomic and therapeutic levels [24], graph representation learning has also been used to support medical diagnosis through the representation of patient records as graphs by using information including brain electrical activity, functional connectivity and anatomical structures [21]. As demonstrated throughout this review, graph-based deep learning has been successfully used to capture phenotypical and topological distributions in histopathology to better enable precision medicine.…”

Section: Discussion and Open Challengesmentioning

confidence: 99%

“…interpretability of the analysis of anatomical structures and microscopic samples. Graph convolutional networks (GCNs) are a deep learningbased method that operate over graphs, and are becoming increasingly useful for medical diagnosis and analysis [21]. GCNs can better exploit irregular relationships and preserve neighboring relations compared with CNN-based models [17].…”

Section: Application #Applications Referencementioning

confidence: 99%

“…With the development of GNNs, deep learning methods have also been applied to GNNs, such as graph convolutional networks (GCNs), attention mechanisms and recurrent neural networks. These frameworks have been proposed as a powerful tool to model functional and anatomical structures, brain electrical activity, and segmentation of the vasculature system and organs [21].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Survey on Graph-Based Deep Learning for Computational Histopathology

Ahmedt-Aristizabal¹,

Armin²,

Denman³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

With the remarkable success of representation learning for prediction problems, we have witnessed a rapid expansion of the use of machine learning and deep learning for the analysis of digital pathology and biopsy image patches. However, traditional learning over patch-wise features using convolutional neural networks limits the model when attempting to capture global contextual information. The phenotypical and topological distribution of constituent histological entities play a critical role in tissue diagnosis. As such, graph data representations and deep learning have attracted significant attention for encoding tissue representations, and capturing intra-and inter-entity level interactions. In this review, we provide a conceptual grounding of graph-based deep learning and discuss its current success for tumor localization and classification, tumor invasion and staging, image retrieval, and survival prediction. We provide an overview of these methods in a systematic manner organized by the graph representation of the input image including whole slide images and tissue microarrays. We also outline the limitations of existing techniques, and suggest potential future advances in this domain.

show abstract

Section: Discussion and Open Challengesmentioning

confidence: 99%

Section: Application #Applications Referencementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Survey on Graph-Based Deep Learning for Computational Histopathology

Ahmedt-Aristizabal¹,

Armin²,

Denman³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Graph Neural Networks (GNNs) have achieved state-of-the-art performance in learning over such relational data in various graph-based machine learning tasks, such as node classification, link prediction, and graph classification [9,20,29,61,68,69]. Due to their superior performance, GNNs are now widely used in many applications such as recommendation systems, credit issuing, traffic forecasting, drug discovery, and medical diagnosis [3,7,17,27,37,63].…”

Section: Introductionmentioning

confidence: 99%

GAP: Differentially Private Graph Neural Networks with Aggregation Perturbation

Sajadmanesh¹,

Shamsabadi²,

Bellet³

et al. 2022

Preprint

View full text Add to dashboard Cite

Graph Neural Networks (GNNs) are powerful models designed for graph data that learn node representation by recursively aggregating information from each node's local neighborhood. However, despite their state-of-the-art performance in predictive graph-based applications, recent studies have shown that GNNs can raise significant privacy concerns when graph data contain sensitive information. As a result, in this paper, we study the problem of learning GNNs with Differential Privacy (DP). We propose GAP, a novel differentially private GNN that safeguards the privacy of nodes and edges using aggregation perturbation, i.e., adding calibrated stochastic noise to the output of the GNN's aggregation function, which statistically obfuscates the presence of a single edge (edge-level privacy) or a single node and all its adjacent edges (node-level privacy). To circumvent the accumulation of privacy cost at every forward pass of the model, we tailor the GNN architecture to the specifics of private learning. In particular, we first precompute private aggregations by recursively applying neighborhood aggregation and perturbing the output of each aggregation step. Then, we privately train a deep neural network on the resulting perturbed aggregations for any node-wise classification task. A major advantage of GAP over previous approaches is that we guarantee edgelevel and node-level DP not only for training, but also at inference time with no additional costs beyond the training's privacy budget. We theoretically analyze the formal privacy guarantees of GAP using Rényi DP. Empirical experiments conducted over three real-world graph datasets demonstrate that GAP achieves a favorable privacy-accuracy trade-off and significantly outperforms existing approaches.

show abstract

“…Graph Neural Networks (GNNs) have emerged as a powerful tool for learning from graph-structured data, and their popularity has surged due to their ability to achieve impressive performance in a wide range of applications, including social network analysis, drug discovery, recommendation systems, and traffic prediction [2,5,14,22,48]. GNNs excel at learning from the structural connectivity of graphs by iteratively updating node embeddings through information aggregation and transformation from neighboring nodes, making them well-suited for tasks such as node classification, graph classification, and link prediction [7,16,25,46,50,51].…”

Section: Introductionmentioning

confidence: 99%

Locally Private Graph Neural Networks

Sajadmanesh

Gática-Pérez

2021

Proceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security

View full text Add to dashboard Cite

Graph Neural Networks (GNNs) have become a popular tool for learning on graphs, but their widespread use raises privacy concerns as graph data can contain personal or sensitive information. Differentially private GNN models have been recently proposed to preserve privacy while still allowing for effective learning over graph-structured datasets. However, achieving an ideal balance between accuracy and privacy in GNNs remains challenging due to the intrinsic structural connectivity of graphs. In this paper, we propose a new differentially private GNN called ProGAP that uses a progressive training scheme to improve such accuracy-privacy trade-offs. Combined with the aggregation perturbation technique to ensure differential privacy, ProGAP splits a GNN into a sequence of overlapping submodels that are trained progressively, expanding from the first submodel to the complete model. Specifically, each submodel is trained over the privately aggregated node embeddings learned and cached by the previous submodels, leading to an increased expressive power compared to previous approaches while limiting the incurred privacy costs. We formally prove that ProGAP ensures edge-level and node-level privacy guarantees for both training and inference stages, and evaluate its performance on benchmark graph datasets. Experimental results demonstrate that ProGAP can achieve up to 5-10% higher accuracy than existing state-of-the-art differentially private GNNs.

show abstract

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

Cited by 4 publications

References 249 publications

A Survey on Graph-Based Deep Learning for Computational Histopathology

A Survey on Graph-Based Deep Learning for Computational Histopathology

GAP: Differentially Private Graph Neural Networks with Aggregation Perturbation

Locally Private Graph Neural Networks

Contact Info

Product

Resources

About