Leveraging clinical data across healthcare institutions for continual learning of predictive risk models

Abstract

The inherent flexibility of machine learning-based clinical predictive models to learn from episodes of patient care at a new institution (site-specific training) comes at the cost of performance degradation when applied to external patient cohorts. To exploit the full potential of cross-institutional clinical big data, machine learning systems must gain the ability to transfer their knowledge across institutional boundaries and learn from new episodes of patient care without forgetting previously learned patterns. In this work, we developed a privacy-preserving learning algorithm named WUPERR (Weight Uncertainty Propagation and Episodic Representation Replay) and validated the algorithm in the context of early prediction of sepsis using data from over 104,000 patients across four distinct healthcare systems. We tested the hypothesis, that the proposed continual learning algorithm can maintain higher predictive performance than competing methods on previous cohorts once it has been trained on a new patient cohort. In the sepsis prediction task, after incremental training of a deep learning model across four hospital systems (namely hospitals H-A, H-B, H-C, and H-D), WUPERR maintained the highest positive predictive value across the first three hospitals compared to a baseline transfer learning approach (H-A: 39.27% vs. 31.27%, H-B: 25.34% vs. 22.34%, H-C: 30.33% vs. 28.33%). The proposed approach has the potential to construct more generalizable models that can learn from cross-institutional clinical big data in a privacy-preserving manner.

Machine learning in transfusion medicine: A scoping review

“…It is crucial to continuously monitor the predictions of a deployed model to ensure that its predictions remain valid and useful. 107 The ability to fine-tune and update models using local 108 and/or more recent data 109 offers huge potential advantage of ML-based predictive modeling over historic simplified scores and static prediction rules in addressing these challenges, enabling models to accommodate new clinical trends and evaluate performance as compared to current practice in an iterative manner. 23 To our knowledge, this review is the first attempt to collate the literature on a wide range of applications of ML in transfusion medicine.…”

Learning across diverse biomedical data modalities and cohorts: Challenges and opportunities for innovation

“… 34 , 35 , 36 This issue persists even when incorporating federated methods due to heterogeneity and lack of local personalization. 37 , 38 …”

Predicting Hospital Readmission among Patients with Sepsis using Clinical and Wearable Data

“…Patient clinical and physiological data during their hospital stay as well as Fitbit data pre-and post-discharge following 90 days after their discharge or until their readmission time were used in our analysis. A total of 40 physiological and clinical variables well established by prior sepsis studies were collected for each patient during their hospitalization [15], [16], [17], [18]. We included pre-discharge and postdischarge wearable data including heart rate, basal metabolic rate calories and activity levels averaged per day.…”