INTERNATIONAL CONGRESS OF
PharmacoGenetics
Converging Genomics and Multi-Modal AI to Advance Precision Medicine
-
Faezeh Arghidash,1,*
1. Department of Medical Biotechnology and Nanotechnology, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran
- Introduction: The past decade has witnessed a rapid evolution in precision medicine driven by advances in genomics and, more recently, by the rise of multi-modal artificial intelligence (AI). Genomic information provides a foundational framework for personalized diagnostics and therapeutic decisions, yet genetic data alone do not fully account for individual health trajectories. By integrating genomics with complementary data streams such as diagnostic imaging, electronic health records (EHR), and social determinants of health multi-modal AI systems hold the promise of more accurate risk stratification, improved disease classification, and more tailored treatment options (1, 2). Real-world applications span oncology, cardiology, and neurology, with several platforms illustrating how tumor genomics, imaging phenotypes, and clinical histories can be combined to guide targeted interventions and monitor outcomes. However, realizing the benefits of this integration requires navigating challenges related to data privacy, governance, and regulatory compliance, model interpretability, and the deployment of learning health systems that can adapt to evolving evidence (3-5). Looking ahead, the field is moving toward broader incorporation of pharmacogenomics, federated data networks, and collaborative frameworks among clinicians, researchers, policymakers, and patients to ensure that advances translate into equitable and effective care.
- Methods: The present study was conducted based on the review of published studies in PubMed, Scopus, and Web of Science, and all studies were reviewed until 2025.
- Results: The integrated framework for precision medicine emerges from the convergence of genomics with multi-modal AI, offering a more holistic approach to personalized care that moves beyond genetics alone to incorporate diverse data streams. By fusing genomic data with imaging, EHR, and other contextual information, multi-modal AI enhances the accuracy of disease risk assessment and classification. Real-world applications span oncology, cardiology, and neurology, with concrete demonstrations showing how tumor genetics, imaging features, and clinical histories can jointly inform treatment choices and monitor patient outcomes. Compared with single modality approaches, cross-modal models tend to achieve better predictive performance, particularly in heterogeneous patient populations. Several challenges accompany these advances: data privacy and security remain critical due to the sensitivity of multi-source data; governance and regulatory compliance must align with standards like general data protection regulation (GDPR) and health insurance portability and accountability act (HIPAA) while enabling collaboration; interpretability is essential so clinicians can understand and trust model recommendations; deploying these systems within learning health networks requires integrating models into continuously evolving clinical workflows and addressing operational hurdles. Looking ahead, the field is moving toward broader adoption of pharmacogenomics, stronger federated data collaborations, and multidisciplinary partnerships to ensure that advances translate into equitable, real-world care.
- Conclusion: In summary, the integration of genomics with multi-modal AI marks a pivotal advance in precision medicine, offering a more comprehensive approach to patient care by weaving together genetic information with diverse data sources such as imaging, EHR, and social determinants of health. This holistic view improves risk stratification, disease classification, and the tailoring of treatments, with real-world applications demonstrating benefits across oncology, cardiology, and neurology. However, realizing these gains requires navigating critical challenges, including safeguarding data privacy and security, aligning governance with regulatory standards, ensuring model interpretability for clinical trust, and embedding learning health systems into dynamic, everyday clinical workflows. Looking forward, the field is poised to expand pharmacogenomics, adopt federated data collaborations, and cultivate multidisciplinary partnerships that translate innovations into equitable, effective care for diverse patient populations.
- Keywords: Precision medicine, Genomics, Artificial intelligence, Pharmacogenomics, Data privacy and security.
