Scalable AI Solutions for IoT-based Healthcare Systems using Cloud Platforms

Abstract

This research offers a solution for AI in IoT-based health care systems that can help large scale implementation of AI-based cloud platforms for real-time data processing and analysis. The IoT sensors with the cloud-enabled environment supported by the CNNs and LSTM networks for health condition monitoring allows real-time management of patient health status. Based on the experimental context, the system was able to prove that the system can actually filter large data stream and work even through the fluctuating network environment. Specifically, the performance targets such as the time taken to process the data, AI output accuracy, the ability of the system to scale and the robustness of the system were assessed. The results demonstrate that the proposed solution has accuracy of up to 96% in health predictions and is easily scalable in terms of the number of IoT devices involved. Network latency was also well controlled, whereby the lowest network latency for networks was measured in 5G networks. Such conclusions prove that the application of cloud-based AI systems can enhance the conditions of the healthcare provisioning, in terms of timely identification of patients with certain health conditions.