Evaluating Medical Lexical Simplification: Rule-Based vs. BERT.

Lexical simplification (LS) can decrease the communication gap between medical experts and laypeople by replacing medical terms with layperson counterparts. In this paper, we present: 1) a rule-based approach to LS using a consumer health vocabulary, and 2) an unsupervised approach using BERT to generate word candidates. Human evaluation shows that the unsupervised model performed better for simplicity and grammaticality, while the rule-based method was better at meaning preservation.

Enabling protocol-based medical critiquing.

This paper investigates the combination of expert critiquing systems and formal medical protocols. Medical protocols might serve as a suitable basis for an expert critiquing system because of the ongoing acceptance of medical protocols and the rise of both evidence-based practice and evidence-based protocols. A prerequisite for a critiquing system based on medical protocols is the ability to match the actions a physician performs in practice to actions prescribed by a protocol. Previous research has shown that this is quite difficult, due to the fact that computerized systems are unable to handle deviations from a protocol, which are common in the medical domain. Our solution to this problem is based on extracting the intention underlying a physician's action and uses the intention as the basis for matching performed actions to prescribed actions. We propose an algorithm for the intention-based matching process and we evaluate the matching algorithm on 12 cases of hyperbilirubinemia in healthy term newborns.