Artificial Intelligence in Preclinical Modeling Klinik Öncesi Modellemede Yapay Zeka
Nuclear Medicine Seminars, cilt.12, sa.2, ss.156-164, 2026 (Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 12 Sayı: 2
- Basım Tarihi: 2026
- Doi Numarası: 10.4274/nts.galenos.2026.99609
- Dergi Adı: Nuclear Medicine Seminars
- Derginin Tarandığı İndeksler: Scopus, EMBASE, Health Research Premium Collection (ProQuest)
- Sayfa Sayıları: ss.156-164
- Anahtar Kelimeler: Artificial intelligence, drug development, personalized medicine, preclinical modeling, tumor imaging
- Sağlık Bilimleri Üniversitesi Adresli: Evet
Özet
The inability of preclinical research methods to fully reflect human biology leads to significant challenges in drug development, including high costs, time losses, and ethical concerns. Consequently, regulatory shifts such as the European Parliament's calls to prioritize non-animal methods and the enactment of the Food and Drug Administration Modernization Act 2.0 in the United States have accelerated the transition toward alternative research models. In this context, artificial intelligence (AI) has emerged not merely as a supportive technology, but as a fundamental component of preclinical modeling. Machine learning and deep learning methods provide researchers with substantial advantages in areas such as biomedical data analysis, evaluation of drug candidates, toxicity prediction, and interpretation of imaging data. In particular, computer-based toxicity and drug evaluation systems help identify failing drug candidates at an early stage, thereby accelerating experimental processes and reducing animal use. In addition, digital twin technologies generate virtual human models using personalized clinical and biological data, enabling different treatment scenarios to be tested safely. Organ-on-chip systems, functioning as miniaturized laboratory models of human organs, offer a more realistic means of examining drug effects. Through AI-powered automation systems, these platforms are becoming faster, more scalable, and more efficient. Furthermore, the integrated analysis of genetic, metabolic, and imaging data enables more comprehensive modeling of biological processes and strengthens efforts to predict patient-specific drug responses in advance. Nevertheless, challenges such as data heterogeneity, model reliability, and interpretability remain significant limitations. In conclusion, AI stands out not merely as a tool for accelerating research processes, but as a new scientific paradigm aimed at reducing animal experimentation, developing models that more closely reflect human biology, and supporting personalized medicine applications.