Automated System Organ Class Classification for Pharmacovigilance Using BioBERT and GPT Frameworks on Efavirenz Adverse Events

DOI

10.2991/978-94-6239-628-9_2How to use a DOI?

Keywords

Machine Learning; LLMs; BioBERT; FAERS; Efavirenz; Pharmacovigilance

Abstract

Efavirenz (EFV) is a first-generation non-nucleoside reverse transcriptase inhibitor and is widely prescribed. However, it is associated with diverse adverse events (AEs), especially persistent neuropsychiatric symptoms. We used U.S. Food and Drug Administration Adverse Event Reporting System (FAERS) data from 2013–2024. In FAERS, AEs are coded as MedDRA Preferred Terms (PTs). We mapped PTs to System Organ Classes (SOCs) to analyse temporal patterns and differences by sex and age. Psychiatric events remained common. In contrast, paediatric and elderly groups were consistently under-represented. To reduce reliance on manual mapping and improve scalability, we developed an automated SOC classifier using BioBERT embeddings and logistic regression (LR). We benchmarked it against random forest, XGBoost, and GPTassisted labelling, and we performed external validation. LR achieved the best macro-F1 (0.825). GPT-assisted labelling achieved 82.8% accuracy (κ = 0.819). Overall, this framework supports reliable SOC-level pharmacovigilance and can be extended to narrative-rich safety data.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Siying Liu
AU  - Indu Bala
PY  - 2026
DA  - 2026/03/31
TI  - Automated System Organ Class Classification for Pharmacovigilance Using BioBERT and GPT Frameworks on Efavirenz Adverse Events
BT  - Proceedings of the International Conference on Recent Trends in Intelligent Computing, Manufacturing, and Electronics (rTIME 2025)
PB  - Atlantis Press
SP  - 7
EP  - 17
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6239-628-9_2
DO  - 10.2991/978-94-6239-628-9_2
ID  - Liu2026
ER  -