Conceptualizing Machine Learning for Dynamic Information Retrieval of Electronic Health Record Notes

Sharon Jiang, Shannon Shen, Monica Agrawal, Barbara Lam, Nicholas Kurtzman, Steven Horng, David R. Karger, David Sontag
Proceedings of the 8th Machine Learning for Healthcare Conference, PMLR 219:343-359, 2023.

Abstract

The large amount of time clinicians spend sifting through patient notes and documenting in electronic health records (EHRs) is a leading cause of clinician burnout. By proactively and dynamically retrieving relevant notes during the documentation process, we can reduce the effort required to find relevant patient history. In this work, we conceptualize the use of EHR audit logs for machine learning as a source of supervision of note relevance in a specific clinical context, at a particular point in time. Our evaluation focuses on the dynamic retrieval in the emergency department, a high acuity setting with unique patterns of information retrieval and note writing. We show that our methods can achieve an AUC of 0.963 for predicting which notes will be read in an individual note writing session. We additionally conduct a user study with several clinicians and find that our framework can help clinicians retrieve relevant information more efficiently. Demonstrating that our framework and methods can perform well in this demanding setting is a promising proof of concept that they will translate to other clinical settings and data modalities (e.g., labs, medications, imaging).

Cite this Paper

BibTeX
@InProceedings{pmlr-v219-jiang23a, title = {Conceptualizing Machine Learning for Dynamic Information Retrieval of Electronic Health Record Notes}, author = {Jiang, Sharon and Shen, Shannon and Agrawal, Monica and Lam, Barbara and Kurtzman, Nicholas and Horng, Steven and Karger, David R. and Sontag, David}, booktitle = {Proceedings of the 8th Machine Learning for Healthcare Conference}, pages = {343--359}, year = {2023}, editor = {Deshpande, Kaivalya and Fiterau, Madalina and Joshi, Shalmali and Lipton, Zachary and Ranganath, Rajesh and Urteaga, Iñigo and Yeung, Serene}, volume = {219}, series = {Proceedings of Machine Learning Research}, month = {11--12 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v219/jiang23a/jiang23a.pdf}, url = {https://proceedings.mlr.press/v219/jiang23a.html}, abstract = {The large amount of time clinicians spend sifting through patient notes and documenting in electronic health records (EHRs) is a leading cause of clinician burnout. By proactively and dynamically retrieving relevant notes during the documentation process, we can reduce the effort required to find relevant patient history. In this work, we conceptualize the use of EHR audit logs for machine learning as a source of supervision of note relevance in a specific clinical context, at a particular point in time. Our evaluation focuses on the dynamic retrieval in the emergency department, a high acuity setting with unique patterns of information retrieval and note writing. We show that our methods can achieve an AUC of 0.963 for predicting which notes will be read in an individual note writing session. We additionally conduct a user study with several clinicians and find that our framework can help clinicians retrieve relevant information more efficiently. Demonstrating that our framework and methods can perform well in this demanding setting is a promising proof of concept that they will translate to other clinical settings and data modalities (e.g., labs, medications, imaging).} }
Endnote
%0 Conference Paper %T Conceptualizing Machine Learning for Dynamic Information Retrieval of Electronic Health Record Notes %A Sharon Jiang %A Shannon Shen %A Monica Agrawal %A Barbara Lam %A Nicholas Kurtzman %A Steven Horng %A David R. Karger %A David Sontag %B Proceedings of the 8th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2023 %E Kaivalya Deshpande %E Madalina Fiterau %E Shalmali Joshi %E Zachary Lipton %E Rajesh Ranganath %E Iñigo Urteaga %E Serene Yeung %F pmlr-v219-jiang23a %I PMLR %P 343--359 %U https://proceedings.mlr.press/v219/jiang23a.html %V 219 %X The large amount of time clinicians spend sifting through patient notes and documenting in electronic health records (EHRs) is a leading cause of clinician burnout. By proactively and dynamically retrieving relevant notes during the documentation process, we can reduce the effort required to find relevant patient history. In this work, we conceptualize the use of EHR audit logs for machine learning as a source of supervision of note relevance in a specific clinical context, at a particular point in time. Our evaluation focuses on the dynamic retrieval in the emergency department, a high acuity setting with unique patterns of information retrieval and note writing. We show that our methods can achieve an AUC of 0.963 for predicting which notes will be read in an individual note writing session. We additionally conduct a user study with several clinicians and find that our framework can help clinicians retrieve relevant information more efficiently. Demonstrating that our framework and methods can perform well in this demanding setting is a promising proof of concept that they will translate to other clinical settings and data modalities (e.g., labs, medications, imaging).
APA
Jiang, S., Shen, S., Agrawal, M., Lam, B., Kurtzman, N., Horng, S., Karger, D.R. & Sontag, D.. (2023). Conceptualizing Machine Learning for Dynamic Information Retrieval of Electronic Health Record Notes. Proceedings of the 8th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 219:343-359 Available from https://proceedings.mlr.press/v219/jiang23a.html.

Related Material