One of the main issues with MANET is Topology control arising due to its dynamic nature. Efficient control of topology in MANET is possible only when mobility prediction is done to avoid any kind of interruption in the communication. In this work, neural network based mobility prediction model for topology control is proposed. In this prediction model a highly scalable and accurate multi-layer architecture is exploited to perform N-step prediction to forecast the future location of the mobile nodes. The optimal path is then selected based on the minimum interference, transmission power values of nodes and the path availability using Ant Colony Optimization (ACO) technique. Clinical care data is collected during the course of ongoing patient care. The proposed method provides uninterrupted communication for transferring clinical care data like details about rehabilitation hospitals for patients. The simulation results obtained prove that the proposed technique is successful in reducing the packet drop, transmission delay and improves the packet delivery ratio as well as residual energy.