Pharmacokinetics of therapeutic information in artificial intelligence: theoretical basis and an ADME model
Vol 20 No 2 (2026), Статьи
Vol 20 No 2 (2026)

Pharmacokinetics of therapeutic information in artificial intelligence: theoretical basis and an ADME model

Статьи
https://doi.org/10.33250/20.02.179
Published June 19, 2026
M. Ya. Golovenko
V. B. Larionov
pdf (Українська)

Keywords

clinical AI
informational pharmacokinetics
Fi, Clᵢ, ADME analogy
biomedical informatics

Abstract

Modern medicine is increasingly integrating clinical artificial intelligence (AI) systems as active tools for decision support; however, their evaluation is predominantly limited to static performance metrics (accuracy, sensitivity, specificity, and ROC curves). Such parameters do not reflect the dynamics of an algorithm’s lifecycle in a real clinical environment, nor do they account for the time of implementation, the scale of adoption, adaptation processes, or the gradual decline of impact. The aim of the study – to develop a theoretical model of the information pharmacokinetics of clinical AI systems, enabling the description of the dynamics of their clinical impact through a structural analogy with the classical pharmacokinetic ADME system. Using this analogy, we propose to formalize the concepts of informational bioavailability (Fi), informational clearance (Clі), informational volume of distribution (Vі), and informational half-life (t½ᵢ), which allows quantitative evaluation of the lifecycle of an algorithm within the clinical environment. To achieve this objective, a comprehensive methodological approach was applied to assess the possibilities for predictive modeling. The proposed model describes the informational impact of clinical AI as a dynamic process and enables prediction of the temporal characteristics of AI influence, assessment of risks associated with the loss of effectiveness, and integration of kinetic metrics that account for the lifecycle of such systems.

https://doi.org/10.33250/20.02.179
pdf (Українська)

References

1. Gentner D. Structure-mapping: a theoretical framework for analogy. Cognitive Science. 1983. V. 7 (2). P. 155–170. https://doi.org/10.1207/s15516709cog0702_3.
2. Weingartner P. Analogy among systems. Dialectica. 1979. V. 33 (3/4). P. 355–378. https://www.jstor.org/stable/42968426.
3. Головенко Н. Я. Физико-химическая фармакология. Одесса : Астропринт, 2004, С. 720–740.
4. Головенко М. Я., Ларіонов В. Б. Штучний інтелект як об’єкт інформаційної фармакології: межа між допоміжним інструментом і лікувальним фактором. Artificial Intelligence. 2026. T. 31 (1). C. 98–116. https://doi.org/10.15407/jai2026.01.098.
5. Golovenko M. Ya., Larionov V. B. From pharmacological informatics to informational pharmacology: a conceptual shift in therapeutics. International Multidisciplinary Conference on Innovation, Technology and Sustainability. Futurity Proceedings. 2026. No. 6. https://doi.org/10.5281/zeno-do.18719816.
6. Dang A., Arora D., Rane P. Role of digital therapeutics and the changing future of healthcare. J. Family Med. Prim. Care. 2020. V. 9 (5). P. 2207–2213. https://doi.org/10.4103/jfmpc.jfmpc_105_20.
7. Bielik L. Methodology of science: an introduction. Bratislava : Comenius University in Bratislava, 2019. 232 p.
8. Spiegelhalter D. A. Trust in numbers: the pursuit of objectivity in science and public life. Princeton, NJ : Princeton University Press, 2019. 224 p.
9. L. von Bertalanffy. General system theory: foundations, development, applications. New York, NY : George Braziller, 1968. 228 p.
10. Rowland M., Thomas N. T. Clinical pharmacokinetics: concepts and applications. 4th ed. Philadelphia, PA Lippincott Williams & Wilkins, 2011. 480 p.
11. Altman D. G., Royston P. What do we mean by validating a prognostic model? Stat. Med. 2000. V. 19 (4). P. 453–473. https://doi.org/10.1002/(SICI)1097-0258(20000229)19:4<453::AID-SIM350>3.0.CO;2-5.
12. Beyond adoption: a new framework for theorizing and evaluating nonadoption, abandonment, and challenges to the scale‑up, spread, and sustainability of health and care technologies. T. Greenhalgh, J. Wherton, C. Papoutsi et al. J. Med. Internet Res. 2017. V. 19 (11). P. e367. https://doi.org/10.2196/jmir.8775.
13. Hesse M. B. Models and analogies in science. Notre Dame, IN : University of Notre Dame Press,
1966. 184 p.
14. World Health Organization. Ethics and governance of artificial intelligence for health: WHO guidance. Geneva : World Health Organization, 2021. 148 p. ISBN 9789240029200. URL: https://www.who.int/publications/i/item/9789240029200.
15. Shargel L., Wu Pong S., Yu A. B. C. Applied biopharmaceutics & pharmacokinetics. 6th ed. New York, NY : McGraw Hill Education / Medical, 2012. 811 p.
16. Brunton L. L., Hilal Dandan R., Knollmann B. C., editors. Goodman & Gilman’s. The pharmacological basis of therapeutics. 13th ed. New York, NY : McGraw Hill, 2018. 1618 p.
17. The bioavailability of drugs — the current state of knowledge. M. Stielow, A. Witczyńska, N. Kubryń et al. Molecules (Basel, Switzerland). 2023. V. 28 (24). P. 8038. https://doi.org/10.3390/mole-cules28248038.
18. Evaluation of solubility limited absorption as a surrogate to predicting positive food effect of BCS II/ IV drugs. K. Rodriguez Fernandez, J. D. Gómez Mantilla, S. Shukla et al. Clin. Pharmacokinet. 2025. V. 64 (3). P. 373–385. https://doi.org/10.1007/s40262-025-01473-9.
19. A systematic review of AI based clinical decision support systems: from development and implementation to applications. S. U. Hasan, S. Hasan, A. Shiwlani et al. Pak. J. Life Soc. Sci. 2025. V. 23 (1). P. 1482–1500. https://doi.org/10.57239/PJLSS 2025 33.1.00115.
20. Artificial intelligence adoption challenges from healthcare providers’ perspectives: a comprehensive review and future directions. M. Abdelwanis, M. C. E. Simsekler, A. F. Gabor et al. Safety Science. 2026. V. 193. P. 107028. https://doi.org/10.1016/j.ssci.2025.107028.
21. Rowland M., Tozer T. N. Clinical pharmacokinetics: concepts and applications. 4th ed. Philadelphia: Lippincott Williams & Wilkins, 1995. 544 p.
22. Toutain P. L., Bousquet-Mélou A. Plasma clearance. J. Vet. Pharmacol. Ther. 2004. V. 27 (6). P. 415–425. https://doi.org/10.1111/j.1365-2885.2004.00605.x
23. Do no harm: a roadmap for responsible machine learning for health care. J. Wiens, S. Saria, M. Sen- dak et al. Nat. Med. 2019. V. 25. P. 1337–1340. https://doi.org/10.1038/s41591-019-0548-6.
24. Key challenges for delivering clinical impact with artificial intelligence. C. J. Kelly, A. Karthikesalingam, M. Suleyman et al. BMC Med. 2019. V. 17 (1). P. 195. https://doi.org/10.1186/s12916-019-1426-2.
25. U. S. Food and Drug Administration. Proposed regulatory framework for modifications to artificial intelligence/machine learning-based software as a medical device (SaMD). Silver Spring, MD : FDA, 2019. URL: https://www.fda.gov/media/122535/download.