English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Foundation models hold promise for transforming AI in healthcare by providingmodular components that are easily adaptable to downstream healthcare tasks,making AI development more scalable and cost-effective. Structured EHRfoundation models, trained on coded medical records from millions of patients,demonstrated benefits including increased performance with fewer traininglabels, and improved robustness to distribution shifts. However, questionsremain on the feasibility of sharing these models across different hospitalsand their performance for local task adaptation. This multi-center studyexamined the adaptability of a recently released structured EHR foundationmodel ($FM_{SM}$), trained on longitudinal medical record data from 2.57MStanford Medicine patients. Experiments were conducted using EHR data at TheHospital for Sick Children and MIMIC-IV. We assessed both adaptability viacontinued pretraining on local data, and task adaptability compared tobaselines of training models from scratch at each site, including a localfoundation model. We evaluated the performance of these models on 8 clinicalprediction tasks. In both datasets, adapting the off-the-shelf $FM_{SM}$matched the performance of GBM models locally trained on all data whileproviding a 13% improvement in settings with few task-specific training labels.With continued pretraining on local data, label efficiency substantiallyimproved, such that $FM_{SM}$ required fewer than 1% of training examples tomatch the fully trained GBM's performance. Continued pretraining was also 60 to90% more sample-efficient than training local foundation models from scratch.Our findings show that adapting shared EHR foundation models across hospitalsprovides improved prediction performance at less cost, underscoring the utilityof base foundation models as modular components to streamline the developmentof healthcare AI.

A Multi-Center Study on the Adaptability of a Shared Foundation Model for Electronic Health Records