Artificial intelligence solutions are used to develop decision support systems in the healthcare field. The developed approaches allow for the creation of artificial patient groups that reflect the fundamental characteristics of the data. Synthetic patient models enable the generation of artificial datasets by mimicking the statistical and clinical characteristics of real patient data. The resulting synthetic datasets offer a significant alternative, especially in situations where accessing large-scale medical datasets is difficult due to restrictions related to patient confidentiality or limitations in data access. Such datasets allow researchers to examine different patient scenarios, simulate treatment processes, and develop AI-based predictive models. Thus, synthetic patient modeling approaches provide a valuable research infrastructure for the development of clinical decision support systems and the investigation of personalized treatment strategies. This study provides an overview of AI-based synthetic patient generation and its potential applications in clinical decision support systems. The study discusses the fundamental components of synthetic patient modeling, including data integration, treatment simulation, dose optimization, and predictive modeling approaches. In addition, the role of advanced machine learning techniques, particularly graph-based deep learning models, in modeling large-scale clinical data and predicting patient-specific treatment outcomes is examined. It highlights how hybrid modeling strategies, combining computational patient simulations with AI algorithms, can support the development of personalized treatment strategies. By integrating simulation-based patient modeling with modern machine learning approaches, it becomes possible to analyze complex treatment dynamics and generate preliminary information that can assist healthcare professionals in clinical decision-making processes. Overall, AI-powered synthetic patient modeling represents a promising research direction for improving data-driven decision support systems and advancing personalized medicine.