FedHealthFog is a federated learning (FL) framework designed for healthcare analytics on fog computing platforms. It addresses the challenges of distributed learning in resource-constrained IoT-enabled healthcare systems, particularly those sensitive to delays and energy efficiency. Traditional FL approaches face issues due to high computational and communication costs, as they rely on a single server for global data aggregation. FedHealthFog improves this by using strategically placed fog nodes as local aggregators within the FL architecture. A greedy heuristic technique optimizes the selection of fog nodes as global aggregators during communication cycles. The framework significantly reduces communication latency (87.01%, 26.90%, 71.74%) and energy consumption (57.98%, 34.36%, 35.37%) compared to benchmark algorithms. Experimental results show that FedHealthFog outperforms existing methods in terms of accuracy, latency, and energy efficiency. The framework reduces the number of global aggregation cycles, enhancing system reliability and performance in delay-sensitive applications. FedHealthFog leverages fog computing to handle resource limitations, reduce latency, save energy, and improve system efficiency in IoT-enabled healthcare systems. The framework's greedy heuristic strategy dynamically selects the best fog node for global aggregation based on workload and latency. The system's performance is evaluated through simulations, demonstrating its effectiveness in reducing communication delays and energy consumption. The results highlight FedHealthFog's potential to transform federated learning in resource-constrained IoT environments for delay-sensitive applications.FedHealthFog is a federated learning (FL) framework designed for healthcare analytics on fog computing platforms. It addresses the challenges of distributed learning in resource-constrained IoT-enabled healthcare systems, particularly those sensitive to delays and energy efficiency. Traditional FL approaches face issues due to high computational and communication costs, as they rely on a single server for global data aggregation. FedHealthFog improves this by using strategically placed fog nodes as local aggregators within the FL architecture. A greedy heuristic technique optimizes the selection of fog nodes as global aggregators during communication cycles. The framework significantly reduces communication latency (87.01%, 26.90%, 71.74%) and energy consumption (57.98%, 34.36%, 35.37%) compared to benchmark algorithms. Experimental results show that FedHealthFog outperforms existing methods in terms of accuracy, latency, and energy efficiency. The framework reduces the number of global aggregation cycles, enhancing system reliability and performance in delay-sensitive applications. FedHealthFog leverages fog computing to handle resource limitations, reduce latency, save energy, and improve system efficiency in IoT-enabled healthcare systems. The framework's greedy heuristic strategy dynamically selects the best fog node for global aggregation based on workload and latency. The system's performance is evaluated through simulations, demonstrating its effectiveness in reducing communication delays and energy consumption. The results highlight FedHealthFog's potential to transform federated learning in resource-constrained IoT environments for delay-sensitive applications.