Learning Explainable Treatment Policies with Clinician-Informed Representations: A Practical Approach

Johannes O Ferstad, Emily B Fox, David Scheinker, Ramesh Johari
Proceedings of the 4th Machine Learning for Health Symposium, PMLR 259:325-349, 2025.

Abstract

Digital health interventions (DHIs) and remote patient monitoring (RPM) have shown great potential in improving chronic disease management through personalized care. However, barriers like limited efficacy and workload concerns hinder adoption of existing DHIs, and limited sample sizes and lack of interpretability limit the effectiveness and adoption of purely black-box algorithmic DHIs. In this paper, we address these challenges by developing a pipeline for learning explainable treatment policies for RPM-enabled DHIs. We apply our approach in the real-world setting of RPM using a DHI to improve glycemic control of youth with type 1 diabetes. Our main contribution is to reveal the importance of \emph{clinical domain knowledge} in developing state and action representations for effective, efficient, and interpretable targeting policies. We observe that policies learned from clinician-informed representations are significantly more efficacious and efficient than policies learned from black-box representations. This work emphasizes the importance of collaboration between ML researchers and clinicians for developing effective DHIs in the real world.

Cite this Paper


BibTeX
@InProceedings{pmlr-v259-ferstad25a, title = {Learning Explainable Treatment Policies with Clinician-Informed Representations: A Practical Approach}, author = {Ferstad, Johannes O and Fox, Emily B and Scheinker, David and Johari, Ramesh}, booktitle = {Proceedings of the 4th Machine Learning for Health Symposium}, pages = {325--349}, year = {2025}, editor = {Hegselmann, Stefan and Zhou, Helen and Healey, Elizabeth and Chang, Trenton and Ellington, Caleb and Mhasawade, Vishwali and Tonekaboni, Sana and Argaw, Peniel and Zhang, Haoran}, volume = {259}, series = {Proceedings of Machine Learning Research}, month = {15--16 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v259/main/assets/ferstad25a/ferstad25a.pdf}, url = {https://proceedings.mlr.press/v259/ferstad25a.html}, abstract = {Digital health interventions (DHIs) and remote patient monitoring (RPM) have shown great potential in improving chronic disease management through personalized care. However, barriers like limited efficacy and workload concerns hinder adoption of existing DHIs, and limited sample sizes and lack of interpretability limit the effectiveness and adoption of purely black-box algorithmic DHIs. In this paper, we address these challenges by developing a pipeline for learning explainable treatment policies for RPM-enabled DHIs. We apply our approach in the real-world setting of RPM using a DHI to improve glycemic control of youth with type 1 diabetes. Our main contribution is to reveal the importance of \emph{clinical domain knowledge} in developing state and action representations for effective, efficient, and interpretable targeting policies. We observe that policies learned from clinician-informed representations are significantly more efficacious and efficient than policies learned from black-box representations. This work emphasizes the importance of collaboration between ML researchers and clinicians for developing effective DHIs in the real world.} }
Endnote
%0 Conference Paper %T Learning Explainable Treatment Policies with Clinician-Informed Representations: A Practical Approach %A Johannes O Ferstad %A Emily B Fox %A David Scheinker %A Ramesh Johari %B Proceedings of the 4th Machine Learning for Health Symposium %C Proceedings of Machine Learning Research %D 2025 %E Stefan Hegselmann %E Helen Zhou %E Elizabeth Healey %E Trenton Chang %E Caleb Ellington %E Vishwali Mhasawade %E Sana Tonekaboni %E Peniel Argaw %E Haoran Zhang %F pmlr-v259-ferstad25a %I PMLR %P 325--349 %U https://proceedings.mlr.press/v259/ferstad25a.html %V 259 %X Digital health interventions (DHIs) and remote patient monitoring (RPM) have shown great potential in improving chronic disease management through personalized care. However, barriers like limited efficacy and workload concerns hinder adoption of existing DHIs, and limited sample sizes and lack of interpretability limit the effectiveness and adoption of purely black-box algorithmic DHIs. In this paper, we address these challenges by developing a pipeline for learning explainable treatment policies for RPM-enabled DHIs. We apply our approach in the real-world setting of RPM using a DHI to improve glycemic control of youth with type 1 diabetes. Our main contribution is to reveal the importance of \emph{clinical domain knowledge} in developing state and action representations for effective, efficient, and interpretable targeting policies. We observe that policies learned from clinician-informed representations are significantly more efficacious and efficient than policies learned from black-box representations. This work emphasizes the importance of collaboration between ML researchers and clinicians for developing effective DHIs in the real world.
APA
Ferstad, J.O., Fox, E.B., Scheinker, D. & Johari, R.. (2025). Learning Explainable Treatment Policies with Clinician-Informed Representations: A Practical Approach. Proceedings of the 4th Machine Learning for Health Symposium, in Proceedings of Machine Learning Research 259:325-349 Available from https://proceedings.mlr.press/v259/ferstad25a.html.

Related Material