Enhanced Security for IOMT Devices: Deep learning Techniques for Network Intrusion Detection
Gopala Krishnan D, M. Diviya, J. K., M. Subramanian · 2024
The increasing popularity of Internet of Medical Things (IoMT) devices, like wearable sensors, has greatly improved patient care by allowing continuous monitoring and real-time data transfer to the cloud. However, this progress also raises security concerns, as the data sent from these devices can be at risk of tampering or hacking during transmission. Keeping this sensitive medical data safe is crucial for patient safety. To address these issues, a fused CNN and LSTM are developed to check whether the data from wearable IoMT devices has been altered or attacked during transmission to the cloud. The CNN part of the model helps to pull out important features from the data, while the LSTM part looks for patterns to spot any signs of tampering. The model uses two convolutional layers each of which is followed by a MaxPooling layer. Our model, tested on a large dataset, achieved an accuracy of 97.06%, showing its ability to protect the integrity of transmitted medical data.