Generative AI Platforms in Predicting Drug Interactions: Case of Digoxin and Warfarin

Cahya Permana Apriansah; Erna Kristin; Soni Siswanto; Purwantiningsih Purwantiningsih; Agung Endro Nugroho

doi:10.2991/978-94-6239-762-0_29

<Previous Article In Volume

Next Article In Volume>

Generative AI Platforms in Predicting Drug Interactions: Case of Digoxin and Warfarin

Authors

Cahya Permana Apriansah¹, Erna Kristin², Soni Siswanto³, Purwantiningsih Purwantiningsih³, Agung Endro Nugroho³^{, *}

¹Master of Clinical Pharmacy Program, Faculty of Pharmacy, Universitas Gadjah Mada, Yogyakarta, Indonesia

²Department of Pharmacology & Therapy, Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada, Yogyakarta, Indonesia

³Department of Pharmacology and Clinical Pharmacy, Faculty of Pharmacy, Universitas Gadjah Mada, Yogyakarta, Indonesia

^*Corresponding author. Email: nugroho_ae@ugm.ac.id

Corresponding Author

Agung Endro Nugroho

Available Online 4 September 2026.

DOI: 10.2991/978-94-6239-762-0_29 How to use a DOI?
Keywords: Drug Interactions; Validity; Artificial Intelligence; Digoxin; Warfarin
Abstract: Drug interactions are a major concern in clinical therapy due to their potential to affect both efficacy and safety. Medications with a narrow therapeutic index, such as digoxin and warfarin, are especially susceptible. With rapid advancements in technology, generative AI platforms like ChatGPT-4o mini, Copilot, and Gemini have emerged as potential tools for identifying drug interactions. However, their validity compared to established resources like UpToDate Lexidrug remains uncertain. This study aimed to evaluate the performance of the three generative AI platforms in detecting drug interactions, using UpToDate Lexidrug as a reference. An observational, cross-sectional design with comparative analysis was applied, and data were analyzed using descriptive statistics. Results showed that ChatGPT-4o mini achieved sensitivity (0.704), specificity (0.714), positive predictive value or PPV (0.812), negative predictive value or NPV (0.579), and accuracy (0.708). Copilot recorded sensitivity (0.859), specificity (0.312), PPV (0.686), NPV (0.558), and accuracy (0.660). Gemini achieved sensitivity (0.889), specificity (0.429), PPV (0.732), NPV (0.688), and accuracy (0.722). Among the platforms, Gemini obtained the highest sensitivity, NPV, and accuracy, while ChatGPT-4o mini demonstrated the best specificity and PPV. These results show that AI has the potential to support drug interaction screening. However, some outputs were inaccurate or incomplete, so generative AI should not replace healthcare professionals in making clinical decisions.
Copyright: © 2026 The Author(s)
Open Access: Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits any noncommercial use, sharing, 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 you modified the licensed material. You do not have permission under this license to share adapted material derived from this chapter or parts of it.

Download article (PDF)

<Previous Article In Volume

Next Article In Volume>

Volume Title: Proceedings of the 6th International Conference on Bioinformatics, Biotechnology, and Biomedical Engineering (BIOMIC 2025)
Series: Advances in Health Sciences Research
Publication Date: 4 September 2026
ISBN: 978-94-6239-762-0
ISSN: 2468-5739
DOI: 10.2991/978-94-6239-762-0_29 How to use a DOI?
Open Access: Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits any noncommercial use, sharing, 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 you modified the licensed material. You do not have permission under this license to share adapted material derived from this chapter or parts of it.

Cite this article

ris enw bib

TY  - CONF
AU  - Cahya Permana Apriansah
AU  - Erna Kristin
AU  - Soni Siswanto
AU  - Purwantiningsih Purwantiningsih
AU  - Agung Endro Nugroho
PY  - 2026
DA  - 2026/09/04
TI  - Generative AI Platforms in Predicting Drug Interactions: Case of Digoxin and Warfarin
BT  - Proceedings of the 6th International Conference on Bioinformatics, Biotechnology, and Biomedical Engineering (BIOMIC 2025)
PB  - Atlantis Press
SP  - 459
EP  - 473
SN  - 2468-5739
UR  - https://doi.org/10.2991/978-94-6239-762-0_29
DO  - 10.2991/978-94-6239-762-0_29
ID  - Apriansah2026
ER  -

download .riscopy to clipboard