Quantifying Explainers of Graph Neural Networks in Computational Pathology

Jaume, Guillaume; Pati, Pushpak; Bozorgtabar, Behzad; Foncubierta, Antonio; Anniciello, Anna Maria; Feroce, Florinda; Rau, Tilman T.; Thiran, Jean‐Philippe; Gabrani, Maria; Göksel, Orçun

doi:10.1109/cvpr46437.2021.00801

Cited by 67 publications

(55 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, in an off-line step, we employ a discriminative feature attribution technique to measure importance scores ∀v ∈ V towards the classification of each class. Specifically, we use GraphGrad-CAM [15,21], a version of Grad-CAM [24] that can operate with GNNs. Argmax across class-wise node attribution maps from GraphGrad-CAM determines the node labels.…”

Section: Methodsmentioning

confidence: 99%

Learning Whole-Slide Segmentation from Inexact and Incomplete Labels Using Tissue Graphs

Anklin

Pati

Jaume

et al. 2021

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Segmenting histology images into diagnostically relevant regions is imperative to support timely and reliable decisions by pathologists. To this end, computer-aided techniques have been proposed to delineate relevant regions in scanned histology slides. However, the techniques necessitate task-specific large datasets of annotated pixels, which is tedious, time-consuming, expensive, and infeasible to acquire for many histology tasks. Thus, weakly-supervised semantic segmentation techniques are proposed to utilize weak supervision that is cheaper and quicker to acquire. In this paper, we propose SegGini, a weakly supervised segmentation method using graphs, that can utilize weak multiplex annotations, i.e., inexact and incomplete annotations, to segment arbitrary and large images, scaling from tissue microarray (TMA) to whole slide image (WSI). Formally, SegGini constructs a tissue-graph representation for an input histology image, where the graph nodes depict tissue regions. Then, it performs weakly-supervised segmentation via node classification by using inexact image-level labels, incomplete scribbles, or both. We evaluated SegGini on two public prostate cancer datasets containing TMAs and WSIs. Our method achieved state-of-the-art segmentation performance on both datasets for various annotation settings while being comparable to a pathologist baseline.

show abstract

Section: Methodsmentioning

confidence: 99%

Learning Whole-Slide Segmentation from Inexact and Incomplete Labels Using Tissue Graphs

Anklin

Pati

Jaume

et al. 2021

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…ASR * , Accuracy [28] Accuracy [29], Faithfulness [30] ASR * , Data Leakage [31] Group/Individual Fairness [32], [33] Standard Evaluations [34] Inference Time [35], Nodes-Per-Joule [36] are commonly used metrics for evaluating GNN explanations. In specific applications, some metrics based on domain knowledge (e.g., correlated separability [59] in computational pathology) are also used to measure explanations. Research Differences.…”

Section: Trustworthy Gnnsmentioning

confidence: 99%

Trustworthy Graph Neural Networks: Aspects, Methods and Trends

Zhang¹,

Wu²,

Yuan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Graph neural networks (GNNs) have emerged as a series of competent graph learning methods for diverse real-world scenarios, ranging from daily applications like recommendation systems and question answering to cutting-edge technologies such as drug discovery in life sciences and n-body simulation in astrophysics. However, task performance is not the only requirement for GNNs. Performance-oriented GNNs have exhibited potential adverse effects like vulnerability to adversarial attacks, unexplainable discrimination against disadvantaged groups, or excessive resource consumption in edge computing environments. To avoid these unintentional harms, it is necessary to build competent GNNs characterised by trustworthiness. To this end, we propose a comprehensive roadmap to build trustworthy GNNs from the view of the various computing technologies involved. In this survey, we introduce basic concepts and comprehensively summarise existing efforts for trustworthy GNNs from six aspects, including robustness, explainability, privacy, fairness, accountability, and environmental well-being. Additionally, we highlight the intricate cross-aspect relations between the above six aspects of trustworthy GNNs. Finally, we present a thorough overview of trending directions for facilitating the research and industrialisation of trustworthy GNNs.

show abstract

“…Despite their great success, GNNs are generally treated as black-box since their decisions are less understood [145,79], leading to the increasing concerns about the explainability of GNNs. It is hard to fully trust the GNN-based models without interpretation to their predictions, and thus restrict their applications in high-stake scenarios such as clinical diagnose [56,147], legal domains [142] and so on. Hence, it is imperative to develop the explanation techniques for the improved transparency of GNNs.…”

Section: Explainabilitymentioning

confidence: 99%

“…However, there is increasing concerns on the reliability of the GNN-based models as they are treated as the black-box. Althogh several explanation methods of GNN are implemented to digital-pathology tasks [56,147,57], it still remains challenging to develop the explanation methods which align with the domain knowledge of the clinical practitioners.…”

Section: Applicationsmentioning

confidence: 99%

A Survey of Trustworthy Graph Learning: Reliability, Explainability, and Privacy Protection

Wu¹,

Li²,

Yu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep graph learning has achieved remarkable progresses in both business and scientific areas ranging from finance and e-commerce, to drug and advanced material discovery. Despite these progresses, how to ensure various deep graph learning algorithms behave in a socially responsible manner and meet regulatory compliance requirements becomes an emerging problem, especially in risk-sensitive domains. Trustworthy graph learning (TwGL) aims to solve the above problems from a technical viewpoint. In contrast to conventional graph learning research which mainly cares about model performance, TwGL considers various reliability and safety aspects of the graph

show abstract

Quantifying Explainers of Graph Neural Networks in Computational Pathology

Cited by 67 publications

References 32 publications

Learning Whole-Slide Segmentation from Inexact and Incomplete Labels Using Tissue Graphs

Learning Whole-Slide Segmentation from Inexact and Incomplete Labels Using Tissue Graphs

Trustworthy Graph Neural Networks: Aspects, Methods and Trends

A Survey of Trustworthy Graph Learning: Reliability, Explainability, and Privacy Protection

Contact Info

Product

Resources

About