Inducing Causal Structure Applied to Glucose Prediction for T1DM Patients

Ana Esponera, Giovanni Cinà
Proceedings of the Fourth Conference on Causal Learning and Reasoning, PMLR 275:920-946, 2025.

Abstract

Causal abstraction techniques such as Interchange Intervention Training (IIT) have been proposed to infuse neural network with expert knowledge encoded in causal models, but their application to real-world problems remains limited. This article explores the application of IIT in predicting blood glucose levels in Type 1 Diabetes Mellitus (T1DM) patients. The study utilizes an acyclic version of the simglucose simulator approved by the FDA to train a Multi-Layer Perceptron (MLP) model, employing IIT to impose causal relationships. Results show that the model trained with IIT effectively abstracted the causal structure and outperformed the standard one in terms of predictive performance across different prediction horizons (PHs) post-meal. Furthermore, the breakdown of the counterfactual loss can be leveraged to explain which part of the causal mechanism are more or less effectively captured by the model. These preliminary results suggest the potential of IIT in enhancing predictive models in healthcare by effectively complying with expert knowledge.

Cite this Paper

BibTeX
@InProceedings{pmlr-v275-esponera25a, title = {Inducing Causal Structure Applied to Glucose Prediction for T1DM Patients}, author = {Esponera, Ana and Cin\`{a}, Giovanni}, booktitle = {Proceedings of the Fourth Conference on Causal Learning and Reasoning}, pages = {920--946}, year = {2025}, editor = {Huang, Biwei and Drton, Mathias}, volume = {275}, series = {Proceedings of Machine Learning Research}, month = {07--09 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v275/main/assets/esponera25a/esponera25a.pdf}, url = {https://proceedings.mlr.press/v275/esponera25a.html}, abstract = {Causal abstraction techniques such as Interchange Intervention Training (IIT) have been proposed to infuse neural network with expert knowledge encoded in causal models, but their application to real-world problems remains limited. This article explores the application of IIT in predicting blood glucose levels in Type 1 Diabetes Mellitus (T1DM) patients. The study utilizes an acyclic version of the simglucose simulator approved by the FDA to train a Multi-Layer Perceptron (MLP) model, employing IIT to impose causal relationships. Results show that the model trained with IIT effectively abstracted the causal structure and outperformed the standard one in terms of predictive performance across different prediction horizons (PHs) post-meal. Furthermore, the breakdown of the counterfactual loss can be leveraged to explain which part of the causal mechanism are more or less effectively captured by the model. These preliminary results suggest the potential of IIT in enhancing predictive models in healthcare by effectively complying with expert knowledge.} }
Endnote
%0 Conference Paper %T Inducing Causal Structure Applied to Glucose Prediction for T1DM Patients %A Ana Esponera %A Giovanni Cinà %B Proceedings of the Fourth Conference on Causal Learning and Reasoning %C Proceedings of Machine Learning Research %D 2025 %E Biwei Huang %E Mathias Drton %F pmlr-v275-esponera25a %I PMLR %P 920--946 %U https://proceedings.mlr.press/v275/esponera25a.html %V 275 %X Causal abstraction techniques such as Interchange Intervention Training (IIT) have been proposed to infuse neural network with expert knowledge encoded in causal models, but their application to real-world problems remains limited. This article explores the application of IIT in predicting blood glucose levels in Type 1 Diabetes Mellitus (T1DM) patients. The study utilizes an acyclic version of the simglucose simulator approved by the FDA to train a Multi-Layer Perceptron (MLP) model, employing IIT to impose causal relationships. Results show that the model trained with IIT effectively abstracted the causal structure and outperformed the standard one in terms of predictive performance across different prediction horizons (PHs) post-meal. Furthermore, the breakdown of the counterfactual loss can be leveraged to explain which part of the causal mechanism are more or less effectively captured by the model. These preliminary results suggest the potential of IIT in enhancing predictive models in healthcare by effectively complying with expert knowledge.
APA
Esponera, A. & Cinà, G.. (2025). Inducing Causal Structure Applied to Glucose Prediction for T1DM Patients. Proceedings of the Fourth Conference on Causal Learning and Reasoning, in Proceedings of Machine Learning Research 275:920-946 Available from https://proceedings.mlr.press/v275/esponera25a.html.

Related Material