Adaptive causal discovery with ordering-based reinforcement learning-approach for type 2 diabetes mellitus risk factors.
7 affiliations
- 1Institute of Vehicle Information Control and Network Technology, Hubei University of Automotive Technology, No. 167, Checheng West Road, Shiyan, 442002, Hubei, China.
- 2College of Electronic and Electrical Engineering, Lingnan Normal University, No. 29 Cunjin Road, Zhanjiang, 524048, Guangdong, China.
- 3School of Electrical and Information Engineering, Hubei University of Automotive Technology, No. 167, Checheng West Road, Shiyan, 442002, Hubei, China.
- 4Institute of Vehicle Information Control and Network Technology, Hubei University of Automotive Technology, No. 167, Checheng West Road, Shiyan, 442002, Hubei, China. [email protected].
- 5College of Computer Science and Intelligent Education, Lingnan Normal University, No. 29 Cunjin Road, Zhanjiang, 524048, Guangdong, China.
- 6College of Automation and Electrical Engineering, Dalian Jiaotong University, No. 794 Huanghe Road, Dalian, 116028, Liaoning, China.
- 7Harbin No. 1 College Hospital, No. 151 Lot Street, Harbin, 150056, Heilongjiang, China.
Abstract
Background
Causal discovery of type 2 diabetes mellitus (T2DM) risk factors is crucial for revealing underlying disease mechanisms and supporting prevention and treatment. However, clinical risk-factor data are typically high-dimensional, limiting the performance of many conventional causal discovery methods. Causal discovery with ordering-based reinforcement learning (CORL) reduces the search space via variable orderings, but it uniformly treats all variable states during initial state construction, failing to account for the heterogeneous importance of individual risk factors.Methods
To address these issues, this study proposes an adaptive CORL framework (ACORL) tailored for T2DM risk factors. First, the search for a causal graph is reformulated into a variable ordering task, where feature contributions are estimated via an ensemble Random Forest classifier to construct a weighted initial state. Next, the variable ordering is optimized through a reinforcement learning reward mechanism to generate a directed acyclic graph. Subsequently, ordinary least squares linear regression pruning and empirical information entropy estimation are applied to eliminate redundant relations and quantify causal strengths. Finally, comparative experiments were conducted on both synthetic and empirical clinical datasets to evaluate structural accuracy and medical plausibility.Results
Simulations on a synthetic dataset demonstrated that ACORL reduced the structural Hamming distance from 3 to 1 and enhanced the F1 score from 50% to 85.71% relative to CORL. On empirical clinical data, ACORL identified causal relations highly consistent with established medical knowledge, achieving a correct-edge rate of 83.3% on the 13-feature NHANES dataset and 91.6% on the 29-feature NHANES dataset. Furthermore, it matched the baseline performance on the lower-dimensional Pima datasets while requiring significantly fewer training iterations.Conclusions
This study introduces an adaptive reinforcement learning framework for T2DM causal discovery. The proposed ACORL model effectively leverages feature heterogeneity and statistical refinement to discover medically plausible causal relations among risk factors. Consequently, it provides an efficient and reliable tool for epidemiological researchers investigating risk-factor structures in large-scale clinical studies.
Related
- Deep reinforcement learning-driven discovery of a MsbA-targeted small-molecule antibiotic for the treatment of Acinetobacter baumannii infection
- CRISPR Technologies in Type 2 Diabetes: From Mechanistic Insights to Therapeutic Discovery
- RNA Design via Conditioned Flow Matching and Finite-Policy Reinforcement Learning
- MORAD: Reinforcement Learning with Multi-Objective Rewards for RNA Inverse Design
- Functional alignment of protein language models via reinforcement learning
BibTeX
@article{chen2026adaptive,
title = {{Adaptive causal discovery with ordering-based reinforcement learning-approach for type 2 diabetes mellitus risk factors.}},
author = {Bo Chen and Ziyi Yuan and Yufeng Chen and Wenxue Xie and Xiue Gao and Jiangang Hu and Yuan Sang},
journal = {BMC bioinformatics},
year = {2026},
month = aug,
doi = {10.1186/s12859-026-06578-x},
url = {https://europepmc.org/article/MED/42972232},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Adaptive causal discovery with ordering-based reinforcement learning-approach for type 2 diabetes mellitus risk factors.
AU - Bo Chen
AU - Ziyi Yuan
AU - Yufeng Chen
AU - Wenxue Xie
AU - Xiue Gao
AU - Jiangang Hu
AU - Yuan Sang
PY - 2026
DA - 2026/08/05
JO - BMC bioinformatics
DO - 10.1186/s12859-026-06578-x
UR - https://europepmc.org/article/MED/42972232
AB - Background Causal discovery of type 2 diabetes mellitus (T2DM) risk factors is crucial for revealing underlying disease mechanisms and supporting prevention and treatment. However, clinical risk-factor data are typically high-dimensional, limiting the performance of many conventional causal discovery methods. Causal discovery with ordering-based reinforcement learning (CORL) reduces the search space via variable orderings, but it uniformly treats all variable states during initial state construction, failing to account for the heterogeneous importance of individual risk factors.Methods To address these issues, this study proposes an adaptive CORL framework (ACORL) tailored for T2DM risk factors. First, the search for a causal graph is reformulated into a variable ordering task, where feature contributions are estimated via an ensemble Random Forest classifier to construct a weighted initial state. Next, the variable ordering is optimized through a reinforcement learning reward mechanism to generate a directed acyclic graph. Subsequently, ordinary least squares linear regression pruning and empirical information entropy estimation are applied to eliminate redundant relations and quantify causal strengths. Finally, comparative experiments were conducted on both synthetic and empirical clinical datasets to evaluate structural accuracy and medical plausibility.Results Simulations on a synthetic dataset demonstrated that ACORL reduced the structural Hamming distance from 3 to 1 and enhanced the F1 score from 50% to 85.71% relative to CORL. On empirical clinical data, ACORL identified causal relations highly consistent with established medical knowledge, achieving a correct-edge rate of 83.3% on the 13-feature NHANES dataset and 91.6% on the 29-feature NHANES dataset. Furthermore, it matched the baseline performance on the lower-dimensional Pima datasets while requiring significantly fewer training iterations.Conclusions This study introduces an adaptive reinforcement learning framework for T2DM causal discovery. The proposed ACORL model effectively leverages feature heterogeneity and statistical refinement to discover medically plausible causal relations among risk factors. Consequently, it provides an efficient and reliable tool for epidemiological researchers investigating risk-factor structures in large-scale clinical studies.
SN - 1471-2105
ER -