Extended Data Fig. 9: Angiogenesis-related environmental markers predicted by the TEA-graph. | Nature Biomedical Engineering

Extended Data Fig. 9: Angiogenesis-related environmental markers predicted by the TEA-graph.

From: Derivation of prognostic contextual histopathological features from whole-slide images of tumours via graph deep learning

Extended Data Fig. 9: Angiogenesis-related environmental markers predicted by the TEA-graph.The alternative text for this image may have been generated using AI.

a, Kaplan-Meier plot of subgraph cluster zero of high IG group (n = 831). b, Graph-level Kaplan-Meier analysis of selected subgraph related to the angiogenesis in subgraph cluster zero of high IG group (n = 831). c, Edge distribution (connectivity) of subgraph cluster zero and angiogenesis-related connectivity. d, Example of the subgraph-level angiogenesis-related pathological features (left) and connected patch cluster (right). Patch size is 80 μm. e, Additional example of pathological features of subgraphs that had high similarity with selected angiogenesis-related subgraph in graph cluster zero in high IG group. Scale bar, 400 μm. (a-b), P-values were calculated through two-sided log-rank test.

Source data

Back to article page