Deep Learning Based Security Solutions for the Internet of Medical Things

With the latest advances in medical technology, more patients today have implantable embedded medical devices. And many of these devices are “connected”: devices are now connected to each other and to the internet, which leads to the use of the term Internet of Medical Things. WiFi/Cellular chips allow medical devices talk to each other, in addition to the traditional roles of sensing and actuating. But there are also new risks, because this connectivity makes these devices more prone to hacking.

This talk focuses on how deep learning techniques can be utilized to make these devices secure. This talk covers different problems of security ranging between internal to communication attacks such as authentication, classification, prediction on implantable medical devices such as insulin pump implants, deep brain stimulators and cardiac defibrillator. This talk will cover efficient techniques, such as multi-layer perceptron neural networks, recurrent neural networks, etc. to overcome these problems. This talk also discusses how these algorithms can be implemented on the node or on the edge to enable real-time decision making.

Speaker(s): Heena Rathore,

Virtual: https://events.vtools.ieee.org/m/308471