Distantly supervised end-to-end medical entity extraction from electronic health records with human-level quality

TL;DR

This paper introduces an end-to-end transformer-based method for medical entity extraction from electronic health records, achieving human-level accuracy without manual annotation by leveraging large unlabeled datasets and medical knowledge bases.

Contribution

The paper presents a novel single-step, end-to-end approach for medical entity extraction using a fine-tuned transformer model trained with distant supervision, eliminating the need for separate recognition and normalization stages.

Findings

Achieves human-level classification quality for frequent entities.

Performs well with large unlabeled EHR datasets and knowledge bases.

Demonstrates end-to-end medical entity extraction without manual supervision.

Abstract

Medical entity extraction (EE) is a standard procedure used as a first stage in medical texts processing. Usually Medical EE is a two-step process: named entity recognition (NER) and named entity normalization (NEN). We propose a novel method of doing medical EE from electronic health records (EHR) as a single-step multi-label classification task by fine-tuning a transformer model pretrained on a large EHR dataset. Our model is trained end-to-end in an distantly supervised manner using targets automatically extracted from medical knowledge base. We show that our model learns to generalize for entities that are present frequently enough, achieving human-level classification quality for most frequent entities. Our work demonstrates that medical entity extraction can be done end-to-end without human supervision and with human quality given the availability of a large enough amount of unlabeled EHR and a medical knowledge base.