AI-driven cloud security framework for protecting healthcare IoT devices and medical data from cyber threats

The growing integration of Internet of Things (IoT) devices in healthcare networks, i.e., smart ICU beds and wearable monitors, has brought about massive cybersecurity issues. This paper puts forth a cloud-based Transformer model designed and optimized using a Genetic Algorithm for real-time anomaly detection in IoT healthcare networks. The model is tested on the IoT Healthcare Security Dataset, which comprises both benign and malicious traffic behaviours like DoS, spoofing, and unauthorized access. The system proposed was found to exhibit outstanding performance, with an accuracy of 98.35%, precision of 97.89%, recall of 98.12%, and an F1-Score of 98.00%.Extensive testing in normal and attack conditions revealed minimal throughput degradation, with the model utilizing resources effectively, having quick detection latencies, and low cloud processing overhead. These findings indicate that the system can support large-scale IoT healthcare networks with little effect on performance. Compared to existing models, the proposed framework demonstrates a dramatic improvement in threat detection accuracy, which indicates its effectiveness in detecting cyberattacks. In general, the cloud-based Transformer model with the Genetic Algorithm provides a scalable, efficient, and highly precise solution for securing healthcare IoT systems. The framework guarantees patient safety, improves system resilience, and supports proactive threat detection and response, making it an effective tool for securing contemporary healthcare environments against changing cyber threats.