What question did this study set out to answer?

This research aims to evaluate the effectiveness of multilingual pre-trained models in clinical report tasks across English and Japanese.

April 1, 2026Open Access

NICTmed at the NCTIR-16 Real-MedNLP Task

Key Points

This research aims to evaluate the effectiveness of multilingual pre-trained models in clinical report tasks across English and Japanese.
Engaged in NTCIR-16 Real-MedNLP challenge focusing on Subtasks 1 and 3 for English and Japanese.
Conducted Named Entity Recognition (NER) and information extraction for adverse drug events (ADE).
Utilized multilingual BERT and XLM-RoBERTa models without external data for experiments.
Achieved similar accuracy levels for Japanese and English using multilingual pre-trained models.
Ranked 3rd in Entity F1 for Subtask1-CR-JA and top rank in Report-level precision and F1 for Subtask3-CR-JA.

Abstract

This paper describes NICTmed team's challenge to Subtask1-CR-EN, Subtask1-CR-JA, Subtask3-CR-EN, and Subtask3-CR-JA in NTCIR-16 Real-MedNLP. In Real-MedNLP, approximately 100 annotated real clinical reports in both English and Japanese are given to participants. Subtask1-CR-EN/JA and Subtask3-CR-EN/JA are both based on Case Reports, Subtask1 is few-resource Named Entity Recognition (NER) and Subtask3 is information extraction for adverse drug event (ADE). We used multilingual BERT (mBERT) and XLM-RoBERTa (XLM-R) to compare how effective the multilingual pre-trained models work in specific domain downstream tasks English and Japanese. Our experiment used no external data to adjust conditions of English and Japanese experiments. As a result, we confirmed that the multilingual pre-training models provide similar level of accuracy in Japanese as in English, and got rank 3 in Entity F1 of all target entities for Subtask1-CR-JA, top rank in Report-level precision and F1 for Subtask3-CR-JA.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper