Artificial intelligence and machine learning in smart vaginal formulation development

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Not applicable — no quantitative effect sizes reported; conclusions are narrative and directional, indicating that AI/ML integration 'can enhance predictability, efficiency, and reproducibility' in vaginal formulation development

• Primary Outcome: Qualitative synthesis of the current and potential role of AI/ML in vaginal drug delivery formulation design, optimisation, and personalisation

• Nnt Or Sensitivity: Not applicable — no clinical trial data or diagnostic accuracy data synthesised; no NNT, sensitivity, specificity, or hazard ratios calculable from this review

• Confidence Interval: Not reported — no meta-analytic synthesis performed

Clinical Application

Clinical implementation of AI/ML-designed personalised vaginal formulations remains a future aspiration rather than a current clinical reality. Barriers include: limited high-quality training datasets for vaginal microenvironment modelling; regulatory uncertainty regarding AI/ML-assisted pharmaceutical development pathways; manufacturing scalability of personalised formulations; cost and accessibility, particularly in resource-limited settings; and the need for prospective clinical validation of AI/ML-optimised products. Digital health monitoring integration (e.g., vaginal biosensors) adds further technical and regulatory complexity. In Australia, vaginal drug delivery products are regulated by the Therapeutic Goods Administration (TGA) under the Australian Register of Therapeutic Goods (ARTG). The TGA has begun engaging with AI/ML in medical device regulation but has not yet published specific guidance for AI/ML-assisted pharmaceutical formulation development. Relevant PBS-listed vaginal products include clindamycin cream and metronidazole gel for bacterial vaginosis, and various antifungal preparations. The RACGP and RANZCOG guidelines for vaginal infections and menopausal management do not currently incorporate AI/ML-guided formulation selection. Australian women's health research priorities, including Indigenous women's health equity and high rates of bacterial vaginosis in remote communities, represent important population-specific considerations that AI/ML personalisation approaches could theoretically address but have not yet been studied in this context. The review's findings are relevant to Australian pharmaceutical researchers and women's health clinicians as a horizon-scanning resource rather than a guide to immediate practice change. Women across the reproductive lifespan requiring vaginal drug delivery for indications including bacterial vaginosis, vulvovaginal candidiasis, sexually transmitted infection prevention (including HIV pre-exposure prophylaxis), contraception, menopausal hormone therapy, and cervical cancer treatment. Biological variability in vaginal pH, microbiota (particularly Lactobacillus-dominant versus dysbiotic communities), and hormonal milieu are identified as key patient-level factors that AI/ML approaches aim to address.

Abstract

Vaginal drug delivery in women's health remains underutilized and insufficiently studied, largely due to the complexity and dynamic nature of the vaginal microenvironment. Variations in vaginal pH, hormonal levels, and microbiota composition introduce significant biological variability, complicating formulation design and contributing to inconsistent therapeutic outcomes and poor patient adherence. Conventional vaginal formulations often fail to account for these individual differences, highlighting the need for more adaptive and predictive approaches. Emerging advances in artificial intelligence (AI) and machine learning (ML) offer promising strategies to address these challenges by enabling multi-parameter, data-driven formulation development that explicitly considers biological variability. Despite their transformative potential, the application of AI and ML in vaginal drug delivery remains limited. This review addresses this gap by examining the emerging role of AI/ML in the design, optimization, and development of next-generation vaginal formulations, including mucoadhesive, mucus-penetrating, bio-responsive, and controlled-release systems. Key factors influencing formulation performance, such as vaginal microbiota composition, are discussed, and ethical, regulatory, and technical challenges relevant to integrating AI/ML into vaginal drug delivery are also considered. Although research is in its early stages, existing studies indicate that AI and ML integration can enhance predictability, efficiency, and reproducibility in vaginal formulation development. Finally, we highlight the potential of developing personalized medicines through the integration of AI/ML and digital health monitoring devices to create patient-centered vaginal drug delivery systems, thereby advancing therapeutic outcomes and transforming women's health.

References

1. 1.Ogundemuren, D. A., Agrahari, V., Wong, A. P., Herrera, C., Ilomuanya, M. O., & Doncel, G. F. (2026). Artificial intelligence and machine learning in smart vaginal formulation development. Advanced Drug Delivery Reviews. https://doi.org/10.1016/j.addr.2026.115882