Improving ARDS Diagnosis Through Context-Aware Concept Bottleneck Models

Anish Narain, Ritam Majumdar, Nikita Narayanan, Dominic C Marshall, Sonali Parbhoo
Proceedings of the 10th Machine Learning for Healthcare Conference, PMLR 298, 2025.

Abstract

The digitization of medical data has opened the door for AI to improve healthcare delivery, but the opaque nature of AI technologies presents challenges for interpretability, which is crucial in clinical settings. Previous work has attempted to explain predictions using Concept Bottleneck Models (CBMs), which learn interpretable concepts or feature groupings that map to higher-level clinical ideas, such as disease severity, facilitating human evaluation. However, these models often experience performance limitations when the concepts fail to adequately explain or characterize the task. In our study, we demonstrate the importance of incorporating contextual information from clinical notes to improve CBM performance, particularly in characterizing Acute Respiratory Distress Syndrome (ARDS), using data from MIMIC-IV. Our approach leverages a Large Language Model (LLM) to process clinical notes and generate additional concepts, boosting accuracy by up to 10% compared to existing methods. This method also enables learning more comprehensive concepts, reducing the risk of information leakage and reliance on spurious shortcuts, thus improving the characterization of ARDS.

Cite this Paper

BibTeX
@InProceedings{pmlr-v298-narain25a, title = {Improving {ARDS} Diagnosis Through Context-Aware Concept Bottleneck Models}, author = {Narain, Anish and Majumdar, Ritam and Narayanan, Nikita and Marshall, Dominic C and Parbhoo, Sonali}, booktitle = {Proceedings of the 10th Machine Learning for Healthcare Conference}, year = {2025}, editor = {Agrawal, Monica and Deshpande, Kaivalya and Engelhard, Matthew and Joshi, Shalmali and Tang, Shengpu and Urteaga, Iñigo}, volume = {298}, series = {Proceedings of Machine Learning Research}, month = {15--16 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v298/main/assets/narain25a/narain25a.pdf}, url = {https://proceedings.mlr.press/v298/narain25a.html}, abstract = {The digitization of medical data has opened the door for AI to improve healthcare delivery, but the opaque nature of AI technologies presents challenges for interpretability, which is crucial in clinical settings. Previous work has attempted to explain predictions using Concept Bottleneck Models (CBMs), which learn interpretable concepts or feature groupings that map to higher-level clinical ideas, such as disease severity, facilitating human evaluation. However, these models often experience performance limitations when the concepts fail to adequately explain or characterize the task. In our study, we demonstrate the importance of incorporating contextual information from clinical notes to improve CBM performance, particularly in characterizing Acute Respiratory Distress Syndrome (ARDS), using data from MIMIC-IV. Our approach leverages a Large Language Model (LLM) to process clinical notes and generate additional concepts, boosting accuracy by up to 10% compared to existing methods. This method also enables learning more comprehensive concepts, reducing the risk of information leakage and reliance on spurious shortcuts, thus improving the characterization of ARDS.} }
Endnote
%0 Conference Paper %T Improving ARDS Diagnosis Through Context-Aware Concept Bottleneck Models %A Anish Narain %A Ritam Majumdar %A Nikita Narayanan %A Dominic C Marshall %A Sonali Parbhoo %B Proceedings of the 10th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2025 %E Monica Agrawal %E Kaivalya Deshpande %E Matthew Engelhard %E Shalmali Joshi %E Shengpu Tang %E Iñigo Urteaga %F pmlr-v298-narain25a %I PMLR %U https://proceedings.mlr.press/v298/narain25a.html %V 298 %X The digitization of medical data has opened the door for AI to improve healthcare delivery, but the opaque nature of AI technologies presents challenges for interpretability, which is crucial in clinical settings. Previous work has attempted to explain predictions using Concept Bottleneck Models (CBMs), which learn interpretable concepts or feature groupings that map to higher-level clinical ideas, such as disease severity, facilitating human evaluation. However, these models often experience performance limitations when the concepts fail to adequately explain or characterize the task. In our study, we demonstrate the importance of incorporating contextual information from clinical notes to improve CBM performance, particularly in characterizing Acute Respiratory Distress Syndrome (ARDS), using data from MIMIC-IV. Our approach leverages a Large Language Model (LLM) to process clinical notes and generate additional concepts, boosting accuracy by up to 10% compared to existing methods. This method also enables learning more comprehensive concepts, reducing the risk of information leakage and reliance on spurious shortcuts, thus improving the characterization of ARDS.
APA
Narain, A., Majumdar, R., Narayanan, N., Marshall, D.C. & Parbhoo, S.. (2025). Improving ARDS Diagnosis Through Context-Aware Concept Bottleneck Models. Proceedings of the 10th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 298 Available from https://proceedings.mlr.press/v298/narain25a.html.

Related Material