Mobility data management and analysis have emerged in the last decade as a very active research domain, promoted by academic events (e.g., several dedicated conferences, journal issues, and seminars), international R&D projects (e.g., GeoPKDD, MODAP, MOVE), and industrial initiatives (e.g., the multiple mobility contests that have been organized recently by several organizations). This book documented the richness and significance of the main research achievements in a variety of domains related to mobility data management and showed how several application domains have already benefited from these achievements. It also highlighted two very important areas for new applications related to most advanced technological environments, namely social networks and network sciences. Yet there is much room for further work in all aspects of movement analysis. The following concluding remarks aim at showing some further developments that are expected within the short-term future and that build on the mobility technologies discussed in this book.

Basic Trajectory Framework

A first evidence is that new research projects are needed to expand the scope and coverage of mobility studies, currently mainly restricted to the limited set of basic concepts described in the first chapter of this book. For example, analyzing movement of deforming areas (e.g., floods, pollution clouds, storms, diseases) has received relatively little attention up to now. Yet its economic importance is rapidly increasing as the disastrous effects of natural phenomena linked to current climate change are influencing government policies to promote better analyses of such phenomena.

Introduction

What Is Mobility Data Mining?

The trajectories of a moving object are a powerful summary of its activity related to mobility. As seen in Chapters 3 and 4, such information can be queried in order to retrieve those trajectories (and the objects that own them) that respond to some given search criteria, for instance following a predefined interesting behavior. However, when massive amounts of information are available, we might be able to move a step further and ask that such “interesting behaviors” automatically emerge from the data. That is precisely the domain explored by mobility data mining.

Moving from queries to data mining essentially consists of adding degrees of freedom to the search process that the algorithms perform. For instance, a query might consist of searching those trajectories that at some point perform the following sequence of maneuvers: abrupt slow down, U-turn, and, finally, accelerate. One possible corresponding data mining task, instead, might require one to discover which sequences of maneuvers are performed frequently in the database of trajectories. Then, the output sequences obtained might also contain the slow down → U-turn → accelerate example just mentioned. To perform this data mining process the user needs to specify the general structure of the behaviors he or she searches (sequences), what kind of elements they can contain (the set of maneuvers to consider, as well as a precise way to locate a given maneuver within a trajectory), and a criterion to select “interesting” behaviors – in our example, the user wants only behaviors that appear frequently in the data.

Introduction

In this chapter, we consider the problem of modeling and representing trajectories in the context of database systems. Since about 1995 there has been research on moving objects databases (MODs), also termed spatio-temporal databases. The general goal has been to allow one to represent moving entities in databases and to enable a user to ask all kinds of questions about such movements. This requires extensions of the DBMS data model and query language. Further, DBMS implementation needs to be extended at all levels, for example, by providing data structures for representation of moving objects, efficient algorithms for query operations, indexing and join techniques, extensions of the query optimizer, and extensions of the user interface to visualize and animate moving objects.

Moving objects databases come in two types. The first represents a set of currently moving objects. One is interested in maintaining the current locations and asking queries about current and expected near future locations. The second type maintains complete histories of movement. These are sometimes called trajectory databases and are the topic of this chapter.

Whereas spatio-temporal databases had been around for a much longer time, they supported only discrete changes of geometries over time. The emphasis in the new field of moving objects databases is to consider continuously changing geometries. Neither the position of a car on a road nor the shape and location of a hurricane changes in discrete steps; these are clearly continuous phenomena.

In this chapter, we explore the range of applications for BCI technology. We have already touched upon some medical applications such as restoration of lost motor and sensory function when we examined invasive and noninvasive BCIs in previous chapters. Here we briefly review these applications before exploring applications in other areas such as entertainment, robotic control, gaming, security, and art.

Medical Applications

The field of brain-computer interfacing originated with the goal of helping the paralyzed and the disabled. It is therefore not surprising that some of the major applications of BCIs to date have been in medical technology, particularly restoring sensory and motor function.

Sensory Restoration

One of the most widely used commercial BCIs is the cochlear implant for the deaf,discussed in Section 10.1.1. The cochlear implant is an example of a BCI for sensoryrestoration, as are retinal implants being developed for the blind (Section 10.1.2).

There has not been much research on two other possible types of purely sensoryBCIs, namely, BCIs for somatosensation and BCIs for olfaction and taste. In the caseof the former, the need for a BCI is minimized because it is ot en possible to restoretactile sensation through skin grafting. However, as we saw in Chapter 11, there isconsiderable interest in somatosensory stimulation as a component of bidirectionalBCIs for allowing paralyzed individuals and amputees to, for example, sense objectsbeing grasped or touched by prosthetic devices.

In this chapter, we review the signal-processing methods applied to recorded brain signals in BCIs for tasks ranging from extracting spikes from the raw signals recorded from invasive electrodes to extracting features for classification. For many of the techniques, we use EEG as the noninvasive recording modality to illustrate the concepts involved, although the techniques could be applied to signals from other sources as well such as ECoG and MEG.

Spike Sorting

Invasive approaches to brain-computer interfacing typically rely on recording spikes from an array of microelectrodes. The goal of signal-processing methods for such an input signal is to reliably isolate and extract the spikes being emitted by a single neuron per recording electrode. This procedure is usually called spike sorting.

The signal recorded by an extracellular electrode implanted in the brain is typicallya mixture of signals from several neighboring neurons, with spikes from closerneurons producing larger amplitude del ections in the recorded signal. h is signalis ot en referred to as multiunit hash or neural hash (Figure 4.1A). Although hashcould also potentially be used as input to brain- computer interfaces, the more traditionalform of input is spikes from individual neurons. Spike sorting methods allowspikes from a single neuron to be extracted from hash.

Our brains evolved to control a complex biological device: our body. As we are finding out today, many millennia of evolutionary tinkering has made the brain a surprisingly versatile and adaptive system, to the extent that it can learn to control devices that are radically different from our body. Brain-computer interfacing, the subject of this book, is a new interdisciplinary field that seeks to explore this idea by leveraging recent advances in neuroscience, signal processing, machine learning, and information technology.

The idea of brains controlling devices other than biological bodies has long been a staple of science-fiction novels and Hollywood movies. However, this idea is fast becoming a reality: in the past decade, rats have been trained to control the delivery of a reward to their mouths, monkeys have moved robotic arms, and humans have controlled cursors and robots, all directly through brain activity.

What aspects of neuroscience research have made these advances possible? Whatare the techniques in computing and machine learning that are allowing brains tocontrol machines? What is the current state-of-the-art in brain-computer interfaces(BCIs)? What limitations still need to be overcome to make BCIs more commonplaceand useful for day-to-day use? What are the ethical, moral, and societal implicationsof BCIs? These are some of the questions that this book addresses.

We have thus far focused on BCIs that record signals from the brain and transform those signals to a control signal for an external device. In this chapter, we reverse the direction of control and discuss BCIs that can be used to stimulate and control specific brain circuits. Some of these BCIs have made the transition from the lab to the clinic and are currently being used by human subjects, such as cochlear implants and deep brain stimulators (DBS), while others are still in experimental stages. We divide these BCIs broadly into two classes: BCIs for sensory restoration and BCIs for motor restoration. We also consider the possibility of sensory augmentation.

Restoring Hearing: Cochlear Implants

One of the most successful BCI devices to date is the cochlear implant for restoring or enabling hearing in the deaf. The implant is a good example of how one can convert knowledge of information processing in a neural system, in this case the cochlea, into building a working BCI that can benefit people.

A holy grail of BCI research is to be able to control complex devices using noninvasive recordings of brain signals at high spatial and temporal resolution. Current noninvasive recording techniques capture changes in blood flow or fluctuations in electric/magnetic fields caused by the activity of large populations of neurons, but we are still far from a recording technique that can capture neural activity at the level of spikes noninvasively. In the absence of such a recording technique, researchers have focused on noninvasive techniques such as EEG, MEG, fMRI, and fNIR, and studied how the large-scale population-level brain signals recorded by these techniques can be used for BCI.

Electroencephalographic (EEG) BCIs

The technique of EEG involves recording electrical signals from the scalp (Section 3.1.2). The idea of using EEG to build a BCI was first suggested by Vidal (1973), but progress was limited until the 1990s when the advent of fast and cheap processors sparked a surge of interest in this area, leading to the development of a variety of EEG-based BCI techniques.

Preface

These headlines, from just a few weeks of news stories in 2012, illustrate the growing fascination of the media and the public with the idea of interfacing minds with machines. What is not clear amid all this hype is: (a) What exactly can and cannot be achieved with current brain-computer interfaces (BCIs) (sometimes also called brain-machine interfaces or BMIs)? (b) What techniques and advances in neuroscience and computing are making these BCIs possible? (c) What are the available types of BCIs? and (d) What are their applications and ethical implications? The goal of this book is to answer these questions and provide the reader with a working knowledge of BCIs and BCI techniques.

Among the most important aspects of brain-computer interfacing are ethical issues – issues pertaining to the medical use of BCIs, the use of BCIs for human augmentation and other applications, and the potential for their misuse. Some of these issues fall under the rubric of neuroethics, but other issues are specific to technological aspects of BCIs.

BCI conferences and workshops sometimes include sessions on ethics, and there have been several articles discussing ethical aspects of BCIs and neural interfaces (e.g., Clausen, 2009; Haselager et al., 2009; Tamburrini, 2009; Salvini et al., 2008; Warwick, 2003). However, there are currently no official regulations or guidelines on BCI use, aside from conventional laws regarding medical and legal ethics. As with other technologies in the past, one can expect that as BCIs become more prevalent in society, laws and ethics pertaining to BCI use will likely be codified by medical and governmental regulatory agencies. In the meantime, this chapter surveys the variety of ethical issues and dilemmas surrounding BCI research and BCI use.

Weighing in at about three pounds, the human brain is a marvel of evolutionary engineering. The brain transforms signals from millions of sensors located all over the body into appropriate muscle commands to enact a behavior suitable to the task at hand. This closed-loop, real-time control system remains unsurpassed by any artificially created system despite decades of attempts by computer scientists and engineers.

The brain’s unique information processing capabilities arise from its massively parallel and distributed way of computing. The workhorse of the brain is a type of cell known as a neuron, a complex electrochemical device that receives information from hundreds of other neurons, processes this information, and conveys its output to hundreds of other neurons. Furthermore, the connections between neurons are plastic, allowing the brain’s networks to adapt to new inputs and changing circumstances. This adaptive and distributed mode of computation sets the brain apart from traditional computers, which are based on the von Neumann architecture with a separate central processing unit, memory units, fixed connections between components, and a serial mode of computation.