Efficient Epileptic Seizure Detection Using CNN-Aided Factor Graphs.

We propose a computationally efficient algorithm for seizure detection. Instead of using a purely data-driven approach, we develop a hybrid model-based/data-driven method, combining convolutional neural networks with factor graph inference. On the CHB-MIT dataset, we demonstrate that the proposed method can generalize well in a 6 fold leave-4-patient-out evaluation.

Tabletop molecular communication: text messages through chemical signals.

In this work, we describe the first modular, and programmable platform capable of transmitting a text message using chemical signalling - a method also known as molecular communication. This form of communication is attractive for applications where conventional wireless systems perform poorly, from nanotechnology to urban health monitoring. Using examples, we demonstrate the use of our platform as a testbed for molecular communication, and illustrate the features of these communication systems using experiments.

On-chip molecular communication: analysis and design.

We consider a confined space molecular communication system, where molecules or information carrying particles are used to transfer information on a microfluidic chip. Considering that information-carrying particles can follow two main propagation schemes: passive transport, and active transport, it is not clear which achieves a better information transmission rate. Motivated by this problem, we compare and analyze both propagation schemes by deriving a set of analytical and mathematical tools to measure the achievable information rates of the on-chip molecular communication systems employing passive to active transport.