Michael W. Towsey

Characterization of autonomic dysfunction in patients with irritable bowel syndrome by means of heart rate variability studies

OBJECTIVE:Our aim was to characterize autonomic dysfunction in patients with irritable bowel syndrome (IBS) using heart rate variability (HRV) studies.METHODS:EKG signals were obtained from 35 patients (mean age, 39.1 ± 9.5 yr, M:F ratio = 2.9:1) and 18 healthy controls (mean age, 38.2 ± 6.5 yr, M:F ratio = 2:1) in supine, standing, and deep-breathing modes. Fast Fourier transformation and autoregressive techniques were used to analyze the HRV power spectra in very low (VLF, 0.0078–0.04 Hz), low (LF, 0.04–0.14 Hz), and high (HF, 0.14–0.4 Hz) frequency bands.RESULTS:In the supine position, the VLF power spectral density (PSD) in IBS was significantly higher than normal (3 vs 1.3 beats per minute [bpm]2/Hz, p < 0.01). On changing from the supine to standing position, the normals (NC) had raised median PSDs in the VLF (1.3 vs 12.8 bpm2/Hz, p < 0.01) and LF (1.6 vs 6.1 bpm2/Hz, p < 0.01) bands, as a sign of increased sympathetic tone, whereas the median HF PSDs (parasympathetic tone) remained unchanged (1.8 bpm2/Hz each, p= 0.01p= 0.8). Similarly, the IBS patients had increased VLF (3.04 vs 14.93 bpm2/Hz, p < 0.01) and LF (2.8 vs 8.7 bpm2/Hz, p < 0.01) PSDs on standing up, but the HF PSD was also raised (from 2.4 to 5.7 bpm2/Hz, p= 0.04). On changing from standing to the deep-breathing mode, the normals had a significant increase in the HF (from 1.8 to 10.3 bpm2/Hz, p < 0.001) and a significant reduction of the VLF (from 12.8 to 2.2 bpm2/Hz, p < 0.01) PSDs. The reduction of the LF PSD was not significant (from 6.1 to 5.6 bpm2/Hz, p= 0.6). In IBS, HF PSD remained constant (5.7 bpm2/Hz each, p= 0.6), whereas the LF PSD increased from 8.7 to 24.2 bpm2/Hz (p < 0.0001). The VLF PSD was reduced (from 14.9 to 4.1 bpm2/Hz, p < 0.0001). In IBS, the median sympathovagal outflow ratio was significantly lower in the standing position (1.4 vs 2.8, p < 0.02) and higher in the deep-breathing mode (7.33 vs 0.42, p < 0.0001) than normal.CONCLUSIONS:IBS patients have reduced sympathetic influence on the heart period in response to orthostatic stress and diminished parasympathetic modulation during deep breathing.

Visualization of Long-duration Acoustic Recordings of the Environment

Acoustic recordings of the environment are an important aid to ecologists monitoring biodiversity and environmental health. However, rapid advances in recording technology, storage and computing make it possible to accumulate thousands of hours of recordings, of which, ecologists can only listen to a small fraction. The big-data challenge addressed in this paper is to visualize the content of long-duration audio recordings on multiple scales, from hours, days, months to years. The visualization should facilitate navigation and yield ecologically meaningful information. Our approach is to extract (at one minute resolution) acoustic indices which reflect content of ecological interest. An acoustic index is a statistic that summarizes some aspect of the distribution of acoustic energy in a recording. We combine indices to produce false-color images that reveal acoustic content and facilitate navigation through recordings that are months or even years in duration.

Efficacy of modified backpropagation and optimisation methods on a real-world medical problem

A wide range of modifications to the backpropagation (BP) algorithm, motivated by heuristic arguments and optimisation theory, has been examined on a real-world medical signal classification problem. The method of choice depends both upon the nature of the learning task and whether one wants to optimise learning for speed or generalisation. It was found that, comparatively, standard BP was sufficiently fast and provided good generalisation when the task was to learn the training set within a given error tolerance. However, if the task was to find the global minimum, then standard BP failed to do so within 100000 iterations, but first order methods which adapt the stepsize were as fast as, if not faster than, conjugate gradient and quasi-Newton methods. Second order methods required the same amount of fine tuning of line search and restart parameters as did the first order methods of their parameters in order to achieve optimum performance.

Scaling Acoustic Data Analysis through Collaboration and Automation

Monitoring and assessing environmental health is becoming increasingly important as human activity and climate change place greater pressure on global biodiversity. Acoustic sensors provide the ability to collect data passively, objectively and continuously across large areas for extended periods of time. While these factors make acoustic sensors attractive as autonomous data collectors, there are significant issues associated with large-scale data manipulation and analysis. We present our current research into techniques for analysing large volumes of acoustic data effectively and efficiently. We provide an overview of a novel online acoustic environmental workbench and discuss a number of approaches to scaling analysis of acoustic data, collaboration, manual, automatic and human-in-the loop analysis.

Ecology and acoustics at a large scale

There is likely no silent place on Earth. Sounds produced by inanimate sources, as well as by plants, animals and humans, permeate all environments from the deep-sea to the aerosphere, 24 h a day (Stocker, 2013). Eavesdropping on animal vocal communications has been an active field of research for more than 60 years (Busnel, 1963) and has, for example, alerted us to the possible deleterious effects of anthropogenic noise on animal physiology and behavior (Barber et al., 2011; Brumm and Slabbekoorn, 2005). Such research is the domain of bioacoustics, the study of the emission, propagation and reception of sound by animals (Bradbury and Vehrencamp, 1998). Bioacoustics is necessarily interdisciplinary, with links to ethology, physiology, neurobiology, biomechanics and evolution, but it tends to focus on the acoustic behavior of individuals, groups, or populations almost always with reference to the biological concept of a species. In other words, it is a species-centered discipline. However, species do not live in closed systems— rather they are part of larger hierarchical structures such as guilds, communities, ecosystems and landscapes. Acoustic interactions between species and between species and their environment may impose important constraints on the structure of these higher-level systems and their development through time. The justification for this special edition on ecological acoustics is the need for investigations of the role of acoustics at higher levels of biological organization. Such investigations link bioacoustics to ecology and point to newways of understanding both animal sounds and ecosystem processes. First attempts to link acoustics with ecology involved the development of acoustic diversity indices as indicators of biodiversity (Pieretti et al., 2011; Sueur et al., 2008) andwith the formalization of soundscape ecology (Pijanowski et al., 2011a). Acoustic diversity indices were inspired by classical indices used in biodiversity assessment with the community level as the unit of sampling and analysis. Soundscape ecology is “the study of sound in landscapes based on an understanding of how sound, from various sources — biological, geophysical and anthropogenic, can be used to understand coupled natural–human dynamics across different spatial and temporal scales” (Pijanowski et al, 2011b) (Fig. 1). Scaling up to community or landscape level would help to address three important challenges in ecology: (1) monitoring animal diversity, (2) understanding the interactions between animal species and (3) measuring andmitigating human noise pollution. Monitoring global changes in biodiversity due to urbanization, ecosystem fragmentation, climate change etc. will require techniques that can scale massively. Acoustic monitoring techniques offer this possibility and a key objective of this special edition is to report on acoustic monitoring techniques that can scale. For example, what kind of acoustic-sensing networks are required to monitor oceans, landscapes and cities? Can acoustic diversity indices act as proxies for biodiversity? Can acoustic indices be used to

Towards an Acoustic Environmental Observatory

The need for large scale environmental monitoring to manage environmental change is well established. Ecologists have long used acoustics as a means of monitoring the environment in their field work, and so the value of an acoustic environmental observatory is evident. However, the volume of data generated by such an observatory would quickly overwhelm even the most fervent scientist using traditional methods. In this paper we present our steps towards realising a complete acoustic environmental observatory - i.e. a cohesive set of hardware sensors, management utilities, and analytical tools required for large scale environmental monitoring. Concrete examples of these elements, which are in active use by ecological scientists, are also presented.

Acoustic classification of Australian anurans using syllable features

Acoustic classification of anurans (frogs) has received increasing attention for its promising application in biological and environment studies. In this study, a novel feature extraction method for frog call classification is presented based on the analysis of spectrograms. The frog calls are first automatically segmented into syllables. Then, spectral peak tracks are extracted to separate desired signal (frog calls) from background noise. The spectral peak tracks are used to extract various syllable features, including: syllable duration, dominant frequency, oscillation rate, frequency modulation, and energy modulation. Finally, a k-nearest neighbor classifier is used for classifying frog calls based on the results of principal component analysis. The experiment results show that syllable features can achieve an average classification accuracy of 90.5% which outperforms Mel-frequency cepstral coefficients features (79.0%).

C-Net: a method for generating non-deterministic and dynamic multivariate decision trees

Abstract. Despite the fact that artificial neural networks (ANNs) are universal function approximators, their black box nature (that is, their lack of direct interpretability or expressive power) limits their utility. In contrast, univariate decision trees (UDTs) have expressive power, although usually they are not as accurate as ANNs. We propose an improvement, C-Net, for both the expressiveness of ANNs and the accuracy of UDTs by consolidating both technologies for generating multivariate decision trees (MDTs). In addition, we introduce a new concept, recurrent decision trees, where C-Net uses recurrent neural networks to generate an MDT with a recurrent feature. That is, a memory is associated with each node in the tree with a recursive condition which replaces the conventional linear one. Furthermore, we show empirically that, in our test cases, our proposed method achieves a balance of comprehensibility and accuracy intermediate between ANNs and UDTs. MDTs are found to be intermediate since they are more expressive than ANNs and more accurate than UDTs. Moreover, in all cases MDTs are more compact (i.e., smaller tree size) than UDTs.

Knowledge extraction and recurrent neural networks: an analysis of an Elman network trained on a natural language learning task

We present results of experiments with Elman recurrent neural networks (Elman, 1990) trained on a natural language processing task. The task was to learn sequences of word categories in a text derived from a primary school reader. The grammar induced by the network was made explicit by cluster analysis which revealed both the representations formed during learning and enabled the construction of state-transition diagrams representing the grammar. A network initialised with weights based on a prior knowledge of the texts statistics, learned slightly faster than the original network.

Managing and Analysing Big Audio Data for Environmental Monitoring

Environmental monitoring is becoming critical as human activity and climate change place greater pressures on biodiversity, leading to an increasing need for data to make informed decisions. Acoustic sensors can help collect data across large areas for extended periods making them attractive in environmental monitoring. However, managing and analysing large volumes of environmental acoustic data is a great challenge and is consequently hindering the effective utilization of the big dataset collected. This paper presents an overview of our current techniques for collecting, storing and analysing large volumes of acoustic data efficiently, accurately, and cost-effectively.