GENERATIVE AI FOR PREDICTIVE MODELING IN HEALTHCARE SYSTEMS: A COMPREHENSIVE EVALUATION OF PERFORMANCE AND RELIABILITY

• Dr. Emilia Kuznetsova

  Department of Computer Engineering, Moscow Institute of Electronic Technology, Russia

  Author

This paper investigates the application of Generative Artificial Intelligence (GenAI) models for predictive analytics in healthcare, focusing on disease diagnosis, risk forecasting, and treatment optimization. Recent advancements such as Generative Adversarial Networks (GANs) and Large Language Models (LLMs) are transforming biomedical data interpretation and enabling early detection of critical illnesses. Using a comparative evaluation of three GenAI-based predictive frameworks—GAN-Enhanced Clinical Risk Predictor, Variational Autoencoder (VAE) Diagnostic Model, and LLM-Driven Temporal Health Analyzer—this study assesses accuracy, reliability, robustness, and explainability. Results show that GenAI models significantly outperform traditional machine-learning baselines, achieving improvements of 12–35% across accuracy and recall metrics. While clinical deployment requires improvements in interpretability and data governance, the study highlights GenAI as a viable foundation for next-generation medical decision-support systems.

• pdf

This work is licensed under a Creative Commons Attribution 4.0 International License.