Explainable AI in Pharmacy: Building Trust in AI-Based Drug Recommendation
International Journal of Research in Pharmacy and Allied Science · 2026 · 05 (09) · 147-168
DOI: 10.71431/ijrpas.2026.5909 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Research in Pharmacy and Allied Science, 2026. Verfasst von Mohammed Aabid und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Artificial intelligence (AI) can be found in pharmacy these days and it helps a lot in areas such as drug selection support, personalized treatment, dosage optimization, and the prediction of medication-related risks; the question is: how can these AI-generated suggestions be made comprehensible and reliable to the healthcare professionals? The problem is solved by Explainable Artificial Intelligence (XAI) by supplying explainable features that show the basis of the AI decision. This paper discusses the role of XAI in pharmacy with a special focus on the areas of AI drug recommendations, adverse drug reaction prediction, drug, drug interaction assessment, patient safety, pharmacovigilance, and the interaction between healthcare professional and patients. We reviewed several papers from the literature and found that XAI methods like Local Interpretable Model-Agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP), decision trees, and attention-based approaches have a potential to enhance openness and to enable doctors and nurses to learn about the reasons for the AI predictions and therefore to XAI possibly contribute to more trust, responsibility, clinical judgment, and safe use of AI-assisted prescribing? Despite all the advantages, the challenges are still quite heavy such as the struggle between the desire to have highly accurate prediction and still having explanations understandable to humans, the question of data quality and possible biases, issues of patient confidentiality and data security, absence of common ways to explain things in healthcare, difficulty to connect the system to other tools already in use, the price of the equipment, and the necessity of properly trained doctors and nurses. The next phase should be devoted to the standardized testing of explanations, development of better AI models, combining patient records from different sources, development of legal guidelines, and making the human-AI teamwork efficient. To sum up, XAI offers an opportunity to increase the comprehensibility and usefulness of AI drug recommendation systems while making sure of human control as part of the pharmacy.
Zitieren
APA
Aabid, M., Ahmed, S. K. A. I., Murtuza, A. M. A. M., & Jaweed, T. S. R. (2026). Explainable AI in Pharmacy: Building Trust in AI-Based Drug Recommendation. International Journal of Research in Pharmacy and Allied Science, 05(09), 147-168. https://doi.org/10.71431/ijrpas.2026.5909
Vancouver
Aabid M, Ahmed SKAI, Murtuza AMAM, Jaweed TSR. Explainable AI in Pharmacy: Building Trust in AI-Based Drug Recommendation. International Journal of Research in Pharmacy and Allied Science. 2026;05(09):147-168. doi:10.71431/ijrpas.2026.5909
BibTeX
@article{Aabid2026,
author = {Aabid, Mohammed and Ahmed, Siddiqui Khalid Aziz Imran and Murtuza, Ansari Mohammad Arman Mohammad and Jaweed, Tufail Sayyadain Raza},
title = {Explainable AI in Pharmacy: Building Trust in AI-Based Drug Recommendation},
journal = {International Journal of Research in Pharmacy and Allied Science},
year = {2026},
volume = {05},
number = {09},
pages = {147--168},
doi = {10.71431/ijrpas.2026.5909}
}
EUVIMED