We are building a system to map efficiently between medical device adverse event terminology in the Japanese Federation of Medical Device Associations (JFMDA) and that in the International Medical Device Regulatory Forum (IMDRF) for international harmonization. The purpose of this study was to evaluate the accuracy of machine translation in the IMDRF terminology using deep learning as the one of the steps of constructing the system. We obtained nine models as follows: mBART and m2m-100 (418M parameters and 1.2B parameters) which are multilingual translation models developed by Facebook, GPT-3 developed by Open AI, googletrans developed by Google, our original model created by parallel corpus using IMDRF terminology and Japanese Medical Device Nomenclature code and the three fine-tuned mBART and m2m-100. We evaluated the accuracy of translation in each model using test data extracted from the parallel corpus. Googletrans was the most accurate model for our task, with the highest BLEU score and manual confirmation. GPT-3 was the second accurate model in manual confirmation. The fine-tuned mBART had a slightly higher BLEU score, however after manual confirmation, the quality of translation was significantly reduced. Fine-tuned m2m-100 had lower BLEU score with deterioration in the quality of translation. Our original model had the lowest BLEU due to insufficient amount of data used for training.

　Accurate identity verification is important for utilizing claim information. Identification is essential to reliable research using claim database. In this study, we have developed a new logic that enables accurate identity verification not affected by the change of insurers without the use of name-based IDs. We used the KDB data of Nara Prefecture covering 7 years of health insurance claims for hospitalization, outpatient and DPC. We have generated “New ID”, which is the combination of birth year/month, sex, diagnosis-related codes, and diagnosis dates. The ledger ID based on the insurer ledger held by the Nara Prefectural National Insurance Association was used for the validation of this logic. The personal IDs linked to other personal IDs by the ledger ID and have outpatient claims from the same medical institution were validated. The number of the target ledger IDs was 69,988 and the total number of possible combinations of personal IDs was 9,796,570,300. The number of true positives was 62,643 and the number of false negatives was 7,345. The validation resulted in a sensitivity of 0.90, a specificity of 1.00, a positive predictive value of 0.99, and a negative predictive value of 1.00. In addition, the number of incorrect identification cases was 0.47 per billion. In the future, even if new medical IDs are established, this logic can identify the same individuals not given the new IDs from accumulated claim data.