Skip to main content

Table 3 Comparison of existing graph representation and causal discovery works in biomedicine with heterogeneous patient data

From: Developing a novel causal inference algorithm for personalized biomedical causal graph learning using meta machine learning

Reference

Task

Graph type

Single-task

Multi-task

Rocheteau et al. [19]

outcome prediction

undirected graph

yes

no

Lu et al. [20]

disease prediction

bipartite or undirected graph

yes

no

Li et al. [21]

CTR prediction

weighted fully-connected graph

yes

no

Xu et al. [22]

clinical prediction

hypergraph

yes

no

Zheng et al. [7]

biomedical structure

directed acyclic graph

yes

no

Ours

causal discovery

directed acyclic graph

yes

yes