Logo image
Back
Continually-Adaptive Representation Learning Framework for Time-Sensitive Healthcare Applications
Conference proceeding   Open access

Continually-Adaptive Representation Learning Framework for Time-Sensitive Healthcare Applications

Alberto M Segre, Akash Choudhuri, Hankyu Jang, Philip M Polgreen, Kishlay Jha and Bijaya Adhikari
CIKM '23: Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, Vol.October 2023, pp.4538-4544
CIKM '23: The 32nd ACM International Conference on Information and Knowledge Management (Birmingham, England, United Kingdom, 10/21/2023–10/25/2023)
10/21/2023
DOI: 10.1145/3583780.3615464
PMCID: PMC11920885
PMID: 40110564
Appears in  UI Libraries Support Open Access
url
https://doi.org/10.1145/3583780.3615464View
Published (Version of record) Open Access

Abstract

Continual learning has emerged as a powerful approach to address the challenges of non-stationary environments, allowing machine learning models to adapt to new data while retaining the previously acquired knowledge. In time-sensitive healthcare applications, where entities such as physicians, hospital rooms, and medications exhibit continuous changes over time, continual learning holds great promise, yet its application remains relatively unexplored. This paper aims to bridge this gap by proposing a novel framework, i.e., Continually-Adaptive Representation Learning, designed to adapt representations in response to changing data distributions in evolving healthcare applications. Specifically, the proposed approach develops a continual learning strategy wherein the context information (e.g., interactions) of healthcare entities is exploited to continually identify and retrain the representations of those entities whose context evolved over time. Moreover, different from existing approaches, the proposed approach leverages the valuable patient information present in clinical notes to generate accurate and robust healthcare embeddings. Notably, the proposed continually-adaptive representations have practical benefits in low-resource clinical settings where it is difficult to training machine learning models from scratch to accommodate the newly available data streams. Experimental evaluations on real-world healthcare datasets demonstrate the effectiveness of our approach in time-sensitive healthcare applications such as Clostridioides difficile (C.diff) Infection (CDI) incidence prediction task and medical intensive care unit transfer prediction task.
 clinical notes electronic healthcare records dynamic embeddings continual learning UIOWA OA Agreement

Details

Metrics

16 Record Views
Logo image