Malware Detection Framework Based on Iterative Neighborhood Component Analysis for Internet of Medical Things

The advancement of the medical equipment and its integration with the Internet of Medical Things (IoMT) has facilitated the remote monitoring of the health-related information of the patient and its analytics by the expert. IoMT network generates the large amount of medical data that are remotely accessed by the smart hospital, smart ambulance, and doctors for better observation and monitoring of the patient. In recent years, the smart connectivity of IoMT devices faces various known and unknown malware threats. To protect the IoMT system, there is a need to device an intelligent malware detection system for IoMT network. Thus, this paper proposes a hybrid Deep learning model by integrating the Convolutional Neural Network (CNN) with Iterative Neighbourhood Component Analysis (INCA), for malware detection and identification task, named as (CNN-INCA-MD). The proposed model is evaluated on IoT malware dataset and achieved an accuracy of 96.98% with precision of 96.68%, recall of 96.67% and f1score of 96.66%. The proposed model is robust in detecting malware and outperforms from existing models.