Stephen Marsland

Habituation revisited: an updated and revised description of the behavioral characteristics of habituation.

The most commonly cited descriptions of the behavioral characteristics of habituation come from two papers published almost 40 years ago [Groves, P. M., & Thompson, R. F. (1970). Habituation: A dual-process theory. Psychological Review, 77, 419-450; Thompson, R. F., & Spencer, W. A. (1966). Habituation: A model phenomenon for the study of neuronal substrates of behavior. Psychological Review, 73, 16-43]. In August 2007, the authors of this review, who study habituation in a wide range of species and paradigms, met to discuss their work on habituation and to revisit and refine the characteristics of habituation. This review offers a re-evaluation of the characteristics of habituation in light of these discussions. We made substantial changes to only a few of the characteristics, usually to add new information and expand upon the description rather than to substantially alter the original point. One additional characteristic, relating to long-term habituation, was added. This article thus provides a modern summary of the characteristics defining habituation, and can serve as a convenient primer for those whose research involves stimulus repetition.

A unified information-theoretic approach to groupwise non-rigid registration and model building

The non-rigid registration of a group of images shares a common feature with building a model of a group of images: a dense, consistent correspondence across the group. Image registration aims to find the correspondence, while modelling requires it. This paper presents the theoretical framework required to unify these two areas, providing a groupwise registration algorithm, where the inherently groupwise model of the image data becomes an integral part of the registration process. The performance of this algorithm is evaluated by extending the concepts of generalisability and specificity from shape models to image models. This provides an independent metric for comparing registration algorithms of groups of images. Experimental results on MR data of brains for various pairwise and groupwise registration algorithms is presented, and demonstrates the feasibility of the combined registration/modelling framework, as well as providing quantitative evidence for the superiority of groupwise approaches to registration.

Constructing diffeomorphic representations for the groupwise analysis of nonrigid registrations of medical images

Groupwise nonrigid registrations of medical images define dense correspondences across a set of images, defined by a continuous deformation field that relates each target image in the group to some reference image. These registrations can be automatic, or based on the interpolation of a set of user-defined landmarks, but in both cases, quantifying the normal and abnormal structural variation across the group of imaged structures implies analysis of the set of deformation fields. We contend that the choice of representation of the deformation fields is an integral part of this analysis. This paper presents methods for constructing a general class of multi-dimensional diffeomorphic representations of deformations. We demonstrate, for the particular case of the polyharmonic clamped-plate splines, that these representations are suitable for the description of deformations of medical images in both two and three dimensions, using a set of two-dimensional annotated MRI brain slices and a set of three-dimensional segmented hippocampi with optimized correspondences. The class of diffeomorphic representations also defines a non-Euclidean metric on the space of patterns, and, for the case of compactly supported deformations, on the corresponding diffeomorphism group. In an experimental study, we show that this non-Euclidean metric is superior to the usual ad hoc Euclidean metrics in that it enables more accurate classification of legal and illegal variations.

On-line Novelty Detection for Autonomous Mobile Robots

The use of mobile robots for inspection tasks is an attractive idea. A robot can travel through environments that humans cannot, and can be trained to identify sensor perceptions that signify potential or actual problems without requiring human intervention. However, in many cases, the appearance of a problem can vary widely, and ensuring that the robot does not miss any possible appearance of the problem (false negatives) is virtually impossible using conventional methods. This paper presents an alternative methodology using novelty detection. A neural network is trained to ignore normal perceptions that do not suggest any problems, so that anything that the robot has not sensed before is highlighted as a possible fault. This makes the incidence of false negatives less likely. We propose a novelty filter that can operate on-line, so that each new input is evaluated for novelty with respect to the data seen so far. The novelty filter learns to ignore inputs that have been sensed previously, or where similar inputs have been perceived. We demonstrate the use of the novelty filter on a series of simple inspection tasks using a mobile robot. The robot highlights those parts of an environment that are novel in some way, that is they are not part of the model acquired during exploration of a different environment. We show the effectiveness of the method using inputs from both sonar sensors and a monochrome camera.

A minimum description length objective function for groupwise non-rigid image registration

Non-rigid registration finds a dense correspondence between a pair of images, so that analogous structures in the two images are aligned. While this is sufficient for atlas comparisons, in order for registration to be an aid to diagnosis, registrations need to be performed on a set of images. In this paper, we describe an objective function that can be used for this groupwise registration. We view the problem of image registration as one of learning correspondences from a set of exemplar images (the registration set), and derive a minimum description length (MDL) objective function. We give a brief description of the MDL approach as applied to transmitting both single images and sets of images, and show that the concept of a reference image (which is central to defining a consistent correspondence across a set of images) appears naturally as a valid model choice in the MDL approach. In this paper, we demonstrate both rigid and non-rigid groupwise registration using our MDL objective function on two-dimensional T1 MR images of the human brain, and show that we obtain a sensible alignment. The extension to the multi-modal case is also discussed. We conclude with a discussion as to how the MDL principle can be extended to include other encoding models than those we present here.

Wireless sensor network based smart home: Sensor selection, deployment and monitoring

The ubiquitous nature of miniature wireless sensors and rapid developments in the wireless network technology have revolutionized home monitoring and surveillance systems. The new means and methods of collecting data efficiently and have led to novel applications for indoor wireless sensor networks. The applications are not limited to solely monitoring but can be extended to behavioral recognition. This can be of great value with the elderly as it can allow anomalous behavior to be detected and corrective actions taken accordingly. This paper details the installation and configuration of unobtrusive sensors in an elderly persons house - a smart home in the making - in a small city in New Zealand. The overall system is envisaged to use machine learning to analyze the data generated by the sensor nodes. The novelty of this project is that instead of setting up an artificial test bed of sensors within the University premises, the sensors have been installed in a subjects home so that data can be collected in a real, not artificial, environment.

Using habituation in machine learning

Habituation, a decrement in response to a stimulus that is presented repeatedly without ill effect, can be identified in almost all animals. It can also be used in machine learning to provide a variety of different applications, such as novelty detection, recency encoding, and temporal signal pre-processing. This paper examines how habituation can be mathematically modelled, and discusses how well these models fit the revised characteristics of habituation. It then demonstrates how the models can be combined with neural networks in order to realise the various applications. Finally, some simple experimental results are presented that demonstrate the effectiveness of the methods.

A hamiltonian particle method for diffeomorphic image registration

Diffeomorphic image registration, where images are aligned using diffeomorphic warps, is a popular subject for research in medical image analysis. We introduce a novel algorithm for computing diffeomorphic warps that solves the Euler equations on the diffeomorphism group explicitly, based on a discretisation of the Hamiltonian, rather than using an optimiser. The result is an algorithm that is many times faster than those considered previously.

Behaviour Recognition from Sensory Streams in Smart Environments

One application of smart homes is to take sensor activations from a variety of sensors around the house and use them to recognise the particular behaviours of the inhabitants. This can be useful for monitoring of the elderly or cognitively impaired, amongst other applications. Since the behaviours themselves are not directly observed, only the observations by sensors, it is common to build a probabilistic model of how behaviours arise from these observations, for example in the form of a Hidden Markov Model (HMM). In this paper we present a method of selecting which of a set of trained HMMs best matches the current observations, together with experiments showing that it can reliably detect and segment the sensor stream into behaviours. We demonstrate our algorithm on real sensor data obtained from the MIT PlaceLab. The results show a significant improvement in the recognition accuracy over other approaches.

N -particle dynamics of the Euler equations for planar diffeomorphisms

The Euler equations associated with diffeomorphism groups have received much recent study because of their links with fluid dynamics, computer vision, and mechanics. In this article, we consider the dynamics of N point particles or “blobs” moving under the action of the Euler equations associated with the group of diffeomorphisms of the plane in a variety of different metrics. This dynamical system is already in widespread use in the field of image registration, where the point particles correspond to image landmarks, but its dynamical behavior has not previously been studied. The 2-body problem is always integrable, and we analyze its phase portrait under different metrics. In particular, we show that 2-body capturing orbits (in which the distances between the particles tend to 0 as t  → ∞) can occur when the kernel is sufficiently smooth and the relative initial velocity of the particles is sufficiently large. We compute the dynamics of these “dipoles” with respect to other test particles, and supplement the calculations with simulations for larger N that illustrate the different regimes.