John Yannis Goulermas

Activity identification using body-mounted sensors--a review of classification techniques.

With the advent of miniaturized sensing technology, which can be body-worn, it is now possible to collect and store data on different aspects of human movement under the conditions of free living. This technology has the potential to be used in automated activity profiling systems which produce a continuous record of activity patterns over extended periods of time. Such activity profiling systems are dependent on classification algorithms which can effectively interpret body-worn sensor data and identify different activities. This article reviews the different techniques which have been used to classify normal activities and/or identify falls from body-worn sensor data. The review is structured according to the different analytical techniques and illustrates the variety of approaches which have previously been applied in this field. Although significant progress has been made in this important area, there is still significant scope for further work, particularly in the application of advanced classification techniques to problems involving many different activities.

A Comparison of Feature Extraction Methods for the Classification of Dynamic Activities From Accelerometer Data

Driven by the demands on healthcare resulting from the shift toward more sedentary lifestyles, considerable effort has been devoted to the monitoring and classification of human activity. In previous studies, various classification schemes and feature extraction methods have been used to identify different activities from a range of different datasets. In this paper, we present a comparison of 14 methods to extract classification features from accelerometer signals. These are based on the wavelet transform and other well-known time- and frequency-domain signal characteristics. To allow an objective comparison between the different features, we used two datasets of activities collected from 20 subjects. The first set comprised three commonly used activities, namely, level walking, stair ascent, and stair descent, and the second a total of eight activities. Furthermore, we compared the classification accuracy for each feature set across different combinations of three different accelerometer placements. The classification analysis has been performed with robust subject-based cross-validation methods using a nearest-neighbor classifier. The findings show that, although the wavelet transform approach can be used to characterize nonstationary signals, it does not perform as accurately as frequency-based features when classifying dynamic activities performed by healthy subjects. Overall, the best feature sets achieved over 95% intersubject classification accuracy.

Multivoxel fMRI analysis of color tuning in human primary visual cortex

We use multivoxel pattern analysis (MVPA) to study the spatial clustering of color-selective neurons in the human brain. Our main objective was to investigate whether MVPA reveals the spatial arrangements of color-selective neurons in human primary visual cortex (V1). We measured the distributed fMRI activation patterns for different color stimuli (Experiment 1: cardinal colors (to which the LGN is known to be tuned), Experiment 2: perceptual hues) in V1. Our two main findings were that (i) cone-opponent cardinal color modulations produce highly reproducible patterns of activity in V1, but these were not unique to each color. This suggests that V1 neurons with tuning characteristics similar to those found in LGN are not spatially clustered. (ii) Unique activation patterns for perceptual hues in V1 support current evidence for a spatially clustered hue map. We believe that our work is the first to show evidence of spatial clustering of neurons with similar color preferences in human V1.

Electric Load Forecasting Based on Locally Weighted Support Vector Regression

The forecasting of electricity demand has become one of the major research fields in electrical engineering. Accurately estimated forecasts are essential part of an efficient power system planning and operation. In this paper, a modified version of the support vector regression (SVR) is presented to solve the load forecasting problem. The proposed model is derived by modifying the risk function of the SVR algorithm with the use of locally weighted regression (LWR) while keeping the regularization term in its original form. In addition, the weighted distance algorithm based on the Mahalanobis distance for optimizing the weighting functions bandwidth is proposed to improve the accuracy of the algorithm. The performance of the new model is evaluated with two real-world datasets, and compared with the local SVR and some published models using the same datasets. The results show that the proposed model exhibits superior performance compare to that of LWR, local SVR, and other published models.

New insights into the plantar pressure correlates of walking speed using pedobarographic statistical parametric mapping (pSPM)

This study investigates the relation between walking speed and the distribution of peak plantar pressure and compares a traditional ten-region subsampling (10RS) technique with a new technique: pedobarographic statistical parametric mapping (pSPM). Adapted from cerebral fMRI methodology, pSPM is a digital image processing technique that registers foot pressure images such that homologous structures optimally overlap, thereby enabling statistical tests to be conducted at the pixel level. Following previous experimental protocols, we collected pedobarographic records from 10 subjects walking at three different speeds: slow, normal, and fast. Walking speed was recorded and correlated with the peak pressures extracted from the 10 regions, and subsequently with the peak pixel data extracted after pSPM preprocessing. Both methods revealed significant positive correlation between peak plantar pressure and walking speed over the rearfoot and distal forefoot after Bonferroni correction for multiple comparisons. The 10RS analysis found positive correlation in the midfoot and medial proximal forefoot, but the pixel data exhibited significant negative correlation throughout these regions (p<5x10(-5)). Comparing the statistical maps from the two approaches shows that subsampling may conflate pressure differences evident in pixel-level data, obscuring or even reversing statistical trends. The negative correlation observed in the midfoot implies reduced longitudinal arch collapse with higher walking speeds. We infer that this results from pre- or early-stance phase muscle activity and speculate that preferred walking speed reflects, in part, a balance between the energy required to tighten the longitudinal arch and the apparent propulsive benefits of the stiffened arch.

Path planning in construction sites: performance evaluation of the Dijkstra, A∗, and GA search algorithms

Abstract This paper presents the application of path planning in construction sites according to multiple objectives. It quantitatively evaluates the performance of three optimisation algorithms namely: Dijkstra, A∗, and Genetic algorithms that are used to find multi-criteria paths in construction sites based on transportation and safety-related cost. During a construction project, site planners need to select paths for site operatives and vehicles, which are characterised by short distance, low risks and high visibility. These path evaluation criteria are combined using a multi-objective approach. The criteria can be optimised to present site planners with the shortest path, the safest path, the most visible path or a path that reflects a combination of short distance, low risk and high visibility. The accuracy of the path solutions and the time complexities of the optimisation algorithms are compared and critically analysed.

A Probabilistic Classifier for Transformer Dissolved Gas Analysis With a Particle Swarm Optimizer

This paper presents a Parzen-Windows (PW)-based classifier for transformer fault diagnosis, which is able to interpret transformer dissolved gas analysis (DGA) with a probabilistic scheme. A global optimizer, particle swarm optimizer (PSO), is employed to optimize the parameters of PW to improve fault classification accuracies. First, the essential concept of PW-based classification using PSO is introduced. This probabilistic classification approach is then extended from a simple PW method to classifying fault types on the evidence of various gas ratios. The proposed approach not only allows an intuitive interpretation of the transformer diagnosis, but also provides a DGA reviewer with quantified confidence to support decision making. It can be seen from the results that both the diagnosis accuracy and computational efficiency are improved compared with a number of fault classification techniques.

A dynamic model of the windlass mechanism of the foot: evidence for early stance phase preloading of the plantar aponeurosis

SUMMARY In the present study we have estimated the temporal elongation of the plantar aponeurosis (PA) during normal walking using a subject-specific multi-segment rigid-body model of the foot. As previous studies have suggested that muscular forces at the ankle can pre-load the PA prior to heel-strike, the main purpose of the current study was to test, through modelling, whether there is any tension present in the PA during early stance phase. Reflective markers were attached to bony landmarks to track the kinematics of the calcaneus, metatarsus and toes during barefoot walking. Ultrasonography measurements were performed on three subjects to determine both the location of the origin of the PA on the plantar aspect of the calcaneus, and the radii of the metatarsal heads. Starting with the foot in a neutral, unloaded position, inverse kinematics allowed calculation of the tension in the five slips of the PA during the whole duration of the stance phase. The results show that the PA experienced tension significantly above rest during early stance phase in all subjects (P<0.01), thus providing support for the PA-preloading hypothesis. The amount of preloading and the maximum elongation of the slips of the PA decreased from medial to lateral. The mean maximum tension exerted by the PA was 1.5 BW (body weight) over the three subjects.

The pizzicato knee-joint energy harvester: characterization with biomechanical data and the effect of backpack load

The reduced power requirements of miniaturized electronics offer the opportunity to create devices which rely on energy harvesters for their power supply. In the case of wearable devices, human-based piezoelectric energy harvesting is particularly difficult due to the mismatch between the low frequency of human activities and the high-frequency requirements of piezoelectric transducers. We propose a piezoelectric energy harvester, to be worn on the knee-joint, that relies on the plucking technique to achieve frequency up-conversion. During a plucking action, a piezoelectric bimorph is deflected by a plectrum; when released due to loss of contact, the bimorph is free to vibrate at its resonant frequency, generating electrical energy with the highest efficiency. A prototype, featuring four PZT-5H bimorphs, was built and is here studied in a knee simulator which reproduces the gait of a human subject. Biomechanical data were collected with a marker-based motion capture system while the subject was carrying a selection of backpack loads. The paper focuses on the energy generation of the harvester and how this is affected by the backpack load. By altering the gait, the backpack load has a measurable effect on performance: at the highest load of 24?kg, a minor reduction in energy generation (7%) was observed and the output power is reduced by 10%. Both are so moderate to be practically unimportant. The average power output of the prototype is 2.06???0.3?mW, which can increase significantly with further optimization.

A class boundary preserving algorithm for data condensation

In instance-based machine learning, algorithms often suffer from storing large numbers of training instances. This results in large computer memory usage, long response time, and often oversensitivity to noise. In order to overcome such problems, various instance reduction algorithms have been developed to remove noisy and surplus instances. This paper discusses existing algorithms in the field of instance selection and abstraction, and introduces a new approach, the Class Boundary Preserving Algorithm (CBP), which is a multi-stage method for pruning the training set, based on a simple but very effective heuristic for instance removal. CBP is tested with a large number of datasets and comparatively evaluated against eight of the most successful instance-based condensation algorithms. Experiments showed that our algorithm achieved similar classification accuracies, with much improved storage reduction and competitive execution speeds.