Brain–Machine Interface Engineering
Synthesis Lectures on Biomedical Engineering
iv aBSTRaCT Neural interfaces are one of the most exciting emerging technologies to impact bioengineering and neuroscience because they enable an alternate communication channel linking directly the nervous system with man-made devices. This book reveals the essential engineering principles and signal processing tools for deriving control commands from bioelectric signals in large ensembles of neurons. The topics featured include analysis techniques for determining neural representation,
... g in motor systems, computing with neural spikes, and hardware implementation of neural interfaces. Beginning with an exploration of the historical developments that have led to the decoding of information from neural interfaces, this book compares the theory and performance of new neural engineering approaches for BMIs. KEywoRdS neural interfaces, brain, neural engineering, neuroscience, neural representation, motor systems "Life can only be understood backward, but it must be lived forward." -Soren Kierkegaard What has the past decade taught us about mankind's ability to interface with and read information from the brain? Looking back on our experiences, the salient recollection is how ill-prepared the present theories of microelectronic circuit design and signal processing are for building interfaces and interpreting brain's activity. Although there is plenty of room for future improvement, the combination of critical evaluation of current approaches and a vision of nueroengineering are helping us develop an understanding on how to read the intent of motion in brains. The flow of ideas and discovery conveyed in this book is quite chronological, starting back in 2001 with a multi-university research project lead by Dr. Miguel Nicolelis of Duke University to develop the next-generation BMIs. The series of engineering developments explained in this book were made possible by the collaboration with Miguel, his contagious enthusiasm, vision, and brilliant experimentalism, that have led us in a journey of discovery in new theories for interfacing with the brain. Part of the results presented here also utilize data collected in his laboratory at Duke University. It was also a journey of innovation shared with colleagues in ECE. Dr. John Harris was instrumental in designing the chips and proposing new devices and principles to improve the performance of current devices. Dr. Karl Gugel helped develop the DSP hardware and firmware to create the new generation of portable systems. We were fortunate to count with the intelligence, dedication, and hard work of many students. Dr. Justin Sanchez came on board to link his biomedical knowledge with signal processing, and his stay at University of Florida has expanded our ability to conduct research here. Dr. Sung-Phil Kim painstakingly developed and evaluated the BMI algorithms. Drs. Deniz Erdogmus and Yadu Rao helped with the theory and their insights. Scott Morrison, Shalom Darmanjian, and Greg Cieslewski developed and programmed the first portable systems for online learning of neural data. Later on, our colleagues Dr. Toshi Nishida and Dr. Rizwan Bashirullah open up the scope of the work with electrodes and wireless systems. Now, a second generation of students is leading the push forward; Yiwen Wang, Aysegul Gunduz, Jack DiGiovanna, Antonio Paiva, and Il Park are advancing the scope of the work with spike train Foreword v modeling. This current research taking us to yet another unexplored direction, which is perhaps the best indication of the strong foundations of the early collaboration with Duke. This book is only possible because of the collective effort of all these individuals. To acknowledge appropriately their contributions, each chapter will name the most important players.