With the accelerating urbanization process in contemporary China, metro systems have assumed an increasingly pivotal role within the national transportation infrastructure. Ensuring structural stability of tunnel surrounding rock formations during construction, as well as conducting comprehensive evaluation and accurate prediction of rock deformation patterns, has emerged as an exceptionally critical component of modern tunneling engineering practices. This investigation, conducted within the context of the Chongqing Rail Transit Line 18 Northern Extension Project, specifically examines deformation characteristics in deep-buried large-cross-section tunnels. The research employs four sophisticated time-series prediction models - Convolutional Neural Networks (CNN), Extreme Learning Machines (ELM), Genetic Algorithm-optimized Backpropagation Neural Networks (GA-BP), and Long Short-Term Memory networks (LSTM) to systematically predict both crown settlement deformations and convergence displacements, followed by meticulous comparative analysis of their predictive performance, after which the collapse risk was evaluated using the model with the best results. The results show that the difference between the predicted and measured values of the GA-BP and ELM prediction models is small, while the prediction results of the GA-BP algorithm are slightly better than those of the ELM prediction model, and the prediction effect is better. Furthermore, as excavation progresses temporally, the probability of tunnel collapse manifestation exhibits a corresponding incremental increase. Nevertheless, when the cumulative deformation magnitude stabilizes at equilibrium conditions, the collapse probability invariably remains well within established parameters of operational safety standards.