Bruce Moulton

Unsupervised machine-learning method for improving the performance of ambulatory fall-detection systems

BackgroundFalls can cause trauma, disability and death among older people. Ambulatory accelerometer devices are currently capable of detecting falls in a controlled environment. However, research suggests that most current approaches can tend to have insufficient sensitivity and specificity in non-laboratory environments, in part because impacts can be experienced as part of ordinary daily living activities.MethodWe used a waist-worn wireless tri-axial accelerometer combined with digital signal processing, clustering and neural network classifiers. The method includes the application of Discrete Wavelet Transform, Regrouping Particle Swarm Optimization, Gaussian Distribution of Clustered Knowledge and an ensemble of classifiers including a multilayer perceptron and Augmented Radial Basis Function (ARBF) neural networks.ResultsPreliminary testing with 8 healthy individuals in a home environment yields 98.6% sensitivity to falls and 99.6% specificity for routine Activities of Daily Living (ADL) data. Single ARB and MLP classifiers were compared with a combined classifier. The combined classifier offers the greatest sensitivity, with a slight reduction in specificity for routine ADL and an increased specificity for exercise activities. In preliminary tests, the approach achieves 100% sensitivity on in-group falls, 97.65% on out-group falls, 99.33% specificity on routine ADL, and 96.59% specificity on exercise ADL.ConclusionThe pre-processing and feature-extraction steps appear to simplify the signal while successfully extracting the essential features that are required to characterize a fall. The results suggest this combination of classifiers can perform better than MLP alone. Preliminary testing suggests these methods may be useful for researchers who are attempting to improve the performance of ambulatory fall-detection systems.

Data Clustering Using Variants of Rapid Centroid Estimation

Prior work suggests that particle swarm clustering (PSC) can be a powerful tool for solving clustering problems. This paper reviews parts of the PSC algorithm, and shows how and why a new class of algorithms is proposed in an attempt to improve the efficiency and repeatability of PSC. This new implementation is referred to as rapid centroid estimation (RCE). RCE simplifies the update rules of PSC, and greatly reduces computational complexity by enhancing the efficiency of the particle trajectories. On benchmark evaluations with an artificial dataset that has 80 dimensions and a volume of 5000, the RCE variants have iteration times of less than 0.1 s, which compares to iteration times of 2 s for PSC and modified PSC (mPSC). On UC Irvine (UCI) machine learning benchmark datasets, the RCE variants are much faster than PSC and mPSC, and produce clusters with higher purity and greatly improved optimization speeds. For example, the RCE variants are more than 100 times faster than PSC and mPSC on the UCI breast cancer dataset. It can be concluded that the RCE variants are leaner and faster than PSC and mPSC, and that the new optimization strategies also improve clustering quality and repeatability.

Body-Area-Network transmission power control using variable adaptive feedback periodicity

We propose a class of adaptive power control protocol, where the period between each feedback transmission is adaptively varied to accommodate run-time variation in the quality of each channel. Initial analyses suggest that transmission control protocols with adaptive feedback periodicity can outperform other comparable schemes. For certain measured channels the period can increase to once every few minutes (thousands of packets) and still provides substantial power savings. Adaptive power control protocols also provide the potential to reduce intra-cell interference.

Method for increasing the computation speed of an unsupervised learning approach for data clustering

Clustering can be especially effective where the data is irregular, noisy and/or not differentiable. A major obstacle for many clustering techniques is that they are computationally expensive, hence limited to smaller data volume and dimension. We propose a lightweight swarm clustering solution called Rapid Centroid Estimation (RCE). Based on our experiments, RCE has significantly quickened optimization time of its predecessors, Particle Swarm Clustering (PSC) and Modified Particle Swarm Clustering (mPSC). Our experimental results show that on benchmark datasets, RCE produces generally better clusters compared to PSC, mPSC, K-means and Fuzzy C-means. Compared with K-means and Fuzzy C-means which produces clusters with 62% and 55% purities on average respectively, thyroid dataset has successfully clustered on average 71% purity in 14.3 seconds.

Fast unsupervised learning method for rapid estimation of cluster centroids

Data clustering is a process where a set of data points is divided into groups of similar points. Recent approaches for data clustering have seen the development of unsupervised learning algorithms based on Particle Swarm Optimization (PSO) techniques. These include Particle Swarm Clustering (PSC) and Modified PSC (mPSC) algorithms for solving clustering problems. However, the PSC and mPSC algorithms tend to be computationally expensive when applied to datasets that have higher levels of dimensionality and large volumes. This paper presents a novel and more efficient swarm clustering strategy we call Rapid Centroid Estimation (RCE). We compare the performance of RCE with the performance of PSC and mPSC in several ways including complexity analyses and particle behavior analyses. Our benchmark testing suggests that RCE can reach a solution 274 times quicker than PSC and 270 times quicker than mPSC for a clustering task where the dataset has a dimension of 80 and a volume of 500. We also investigated particle behaviors on two-class two-dimensional datasets with volume of 500, presenting 250 data for each well-separated class with known Gaussian centers. We found that RCE converged to the appropriate centers at 70 updates on average, compared to 19802 updates for PSC and 23006 updates for mPSC. An ANOVA indicates RCE is significantly faster than both PSC and mPSC.

Updating Electronic Health Records with Information from Sensor Systems: Considerations Relating To Standards and Architecture Arising From the Development of a Prototype System

Several countries around the globe are moving towards national and international standards for Electronic Health Records (EHRs). One function of the standards is to guide the long-term convergence of local systems into integrated evolving national health information systems. The Australian commonwealth government is implementing a nationwide EHR system whereby every Australian will be able to upload data to his or her EHR. Thus Australians, if they wish, will eventually be able to upload data from on-body sensors and in-home monitoring systems to their EHRs. This article explores issues associated with the architecture of systems which allow medical records to be updated with information from monitoring/sensor systems. A prototype was developed to determine some of the key architectural considerations. A sensor simulator was implemented for testing purposes which allows a user of the simulator to impersonate a bed or group of in-home or on-body sensors connected with a person who is in a hospital, retirement home or private home. Findings are discussed relating to key architectural considerations including security, maintainability and modularity.

An algorithm for scalable clustering: Ensemble Rapid Centroid Estimation

This paper describes a new algorithm, called Ensemble Rapid Centroid Estimation (ERCE), designed to handle large-scale non-convex cluster optimization tasks, and estimate the number of clusters with quasi-linear complexity. ERCE stems from a recently developed Rapid Centroid Estimation (RCE) algorithm. RCE was originally developed as a lightweight simplification of the Particle Swarm Clustering (PSC) algorithm. RCE retained the quality of PSC, greatly reduced the computational complexity, and increased the stability. However, RCE has certain limitations with respect to complexity, and is unsuitable for non-convex clusters. The new ERCE algorithm presented here addresses these limitations.

Ambulatory health monitoring and remote sensing systems to be used by outpatients and elders at home: User-related design considerations

Recent developments have seen increased interest in the effect of end-user attributes on the in-practice effectiveness of systems that detect incapacitating falls and trauma at home. It is hoped that consideration and evaluation of such issues will ultimately result in long-term benefits including earlier crisis detection and response, reduced hospital admissions, and improved quality of life for relatively large groups of people. Key concerns include the needs and capabilities of end-users, the ability to nominate who is to be alerted, security, privacy, interface design and system failures. It is concluded that particularly relevant avenues for further research include end-user characteristics, interface design and peer-to-peer components.

Optimization strategies for rapid centroid estimation

Particle swarm algorithm has been extensively utilized as a tool to solve optimization problems. Recently proposed particle swarm±based clustering algorithm called the Rapid Centroid Estimation (RCE) is a lightweight alteration to Particle Swarm Clustering (PSC). The RCE in its standard form is shown to be superior to conventional PSC algorithm. We have observed some limitations in RCE including the possibility to stagnate at a local minimum combination and the restriction in swarm size. We propose strategies to optimize RCE further by introducing RCE+ and swarm RCE+. Five benchmark datasets from UCI machine learning database are used to test the performance of these new strategies. In Glass dataset swarm RCE+ is able to achieve highest purity centroid combinations with less iteration (90.3%±1.1% in 9±5 iterations) followed by RCE+ (89%±3.5% in 65±62 iterations) and RCE (87%±5.9% in 54±44). Similar quality is also reflected in other benchmark datasets including Iris, Wine, Breast Cancer, and Diabetes.

Unsupervised segmentation of heel-strike IMU data using rapid cluster estimation of wavelet features

When undertaking gait-analysis, one of the most important factors to consider is heel-strike (HS). Signals from a waist worn Inertial Measurement Unit (IMU) provides sufficient accelerometric and gyroscopic information for estimating gait parameter and identifying HS events. In this paper we propose a novel adaptive, unsupervised, and parameter-free identification method for detection of HS events during gait episodes. Our proposed method allows the device to learn and adapt to the profile of the user without the need of supervision. The algorithm is completely parameter-free and requires no prior fine tuning. Autocorrelation features (ACF) of both antero-posterior acceleration (aAP) and medio-lateral acceleration (aML) are used to determine cadence episodes. The Discrete Wavelet Transform (DWT) features of signal peaks during cadence are extracted and clustered using Swarm Rapid Centroid Estimation (Swarm RCE). Left HS (LHS), Right HS (RHS), and movement artifacts are clustered based on intra-cluster correlation. Initial pilot testing of the system on 8 subjects show promising results up to 84.3%±9.2% and 86.7%±6.9% average accuracy with 86.8%±9.2% and 88.9%±7.1% average precision for the segmentation of LHS and RHS respectively.