Back

Interoperability of standardised electronic healthcare records facilitates transfer learning

Remfry, E.; Henkin, R.

2025-06-12 health informatics
10.1101/2025.06.12.25329419 medRxiv
Show abstract

Electronic healthcare records (EHR) use codes from different vocabularies to describe medical occurrences, often varying by type of care and country. Common data models (CDM) such as the Observational Medical Outcomes Partnership (OMOP) have been developed to enable the combination and comparison of heterogeneous datasets. We use the OMOP Standard Vocabularies to standardise two English EHR datasets and assess their interoperability through a BERT-based transformer model via pretraining and fine-tuning. Our results show the potential for standardisation to empower transfer learning, with tradeoffs related to data loss.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.