In response to the global health crises posed by infectious diseases like COVID-19, this study presents an enhanced SEIR model by introducing a novel feedback control mechanism. This mechanism dynamically adapts not only to the current state of the infected population but also to its rate of change, offering a dual-dependence control strategy. Such an approach significantly enhances the responsiveness and precision of epidemic management interventions, leading to a substantial reduction in peak infection rates and overall disease burden. To achieve optimal control, we employed the gradient descent method for mathematical analysis, ensuring both theoretical robustness and computational efficiency. Theoretical results were validated through comprehensive numerical simulations, demonstrating the efficacy of our control strategy across various epidemic scenarios. Furthermore, a comparative analysis with Pontryagin's maximum principle highlights the superior performance of our model, underscoring the critical role of incorporating both state and rate of change information in designing effective public health interventions. These findings reveal new possibilities for improving epidemic containment strategies, offering valuable insights for real-world disease management.