- Open Access
Unsupervised machine-learning method for improving the performance of ambulatory fall-detection systems
© Yuwono et al; licensee BioMed Central Ltd. 2012
- Received: 12 December 2011
- Accepted: 16 February 2012
- Published: 16 February 2012
Falls 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.
We 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.
Preliminary 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.
The 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.
- Discrete Wavelet Transform
- Radial Basis Function Neural Network
- Fall Detection
- Combine Classifier
- Radial Basis Kernel
Falls are recognised by the World Health Organization as a major cause of hospitalization of older people . If no preventative measures are undertaken, it is estimated that costs associated with fall-related trauma will double over the next 20 years .
Ambulatory accelerometer devices are currently capable of detecting falls in a controlled environment, and these devices are also potentially useful for assessing gait and tremor in older people with Parkinson's disease [2, 3]. Research regarding accelerometer-based fall detection typically uses thresholding algorithms [4–6]. Those algorithms typically determine if a person experiences an acceleration above a certain value - that is, an impact acceleration as a person hits the ground - and sometimes combine this with an approach for measuring whether the impact is followed by a period of lying down/not moving. However, research suggests that those approaches can tend to have limited sensitivity for soft-falls (for example, where a person falls against a wall) and break-falls (for example, where a person reduces the impact of their fall by putting out an arm) [e.g. ]. Current fall-detection approaches can also tend to result in poor specificity, in part because impacts can be experienced as part of ordinary daily living activities.
The fall-detection work undertaken at the University of Technology Sydney is part of a larger research program focused on health technologies that also focuses on issues associated with sensor-transceivers that stream data such as heart rate, electrocardiogram (ECG), oxygen saturation, body temperature, and body position [8–11].
The method described here builds on prior work including the work of Shi and others who combined threshold techniques and Support Vector Machines (SVM) to improve the performance of fall detection classifiers . Our approach attempts to improve the specificity even while only using one sensor device, a waist-worn tri-axial accelerometer.
The feature extraction stage makes use a custom method we refer to as Gaussian distribution of Clustered Knowledge (GCK) signal generation. Clustering is done with Regrouping Particle Swarm Optimization (RegPSO). Classification is undertaken by a newly developed "Augmented Radial Basis Function" (ARBF) neural network  alongside a multilayer perceptron (MLP).
The input data consists of sampled acceleration values in three dimensions. The main accelerometer module used is the RD-3152 MMA7260Q - Zstar2 from Freescale Semiconductor. The module provides 3-axis acceleration data using an MMA7260Q accelerometer set to ± 6 g sensitivity range. Wireless communication is established using a ZigBee protocol 2.4 GHz band to communicate with the receiver board . The accelerometer sensor is placed inside the right pocket of a vest.
At the receiving end of the wireless link, the samples are handled in real time using Java2SE and Matlab. Each signal has a length of 5 seconds sampled with 20Hz sampling frequency. Signals are divided into 2 classes: fall signals and Activities of Daily Living (ADL) signals.
Data preprocessing is undertaken in three steps: uncommon acceleration detection, normalization, and data filtering.
Uncommon acceleration detection: uncommon accelerations, regardless of whether they are due to falls or not, can be observed when the magnitude of acceleration is above a specified threshold. When an uncommon acceleration occurs at time τ, a window is constructed at τ ± 2.5 s and acceleration data in that window is pushed to the classification queue.
Normalization: raw acceleration data has an offset due to the static force of gravity that differs depending on orientation of the accelerometer. Acceleration signals are normalized by subtracting every sample from .
Data Filtering: the DWT decomposes discrete time signals using a digital filter approach. The DWT computes successive convolutions between input signal with discrete low pass and high pass filters . The application of this filter in the system can be seen in Figure 2. The DWT filters the acceleration signal and down-samples it up to the third order using Haar wavelets. The intention is to reduce the signal complexity while still providing sufficient relevant information to the classifiers.
The ARBF function uses a Gaussian radial basis kernel. It can be described as a K-dimensional Gaussian distribution, where K is the number of dimensions of the input. The output of the RBF layer is a vector where μ n and σ n correspond to cluster centroids and the standard deviation of each RBF node. The RBF centroids are optimized using RegPSO. The MLP layer uses a sigmoid kernel in the hidden layer and a linear kernel in the output layer. No normalization method is required at this stage because the RBF layer has already normalized the input signals from 0 to 1. The MLP layer is trained with resilient back-propagation. This combination of MLP and ARBF was used because of each classifier's different characteristics. MLP networks tend to perform better in global generalization, while RBF-kernel based classifiers such as ARBF tend to perform better in local generalization . The ensemble receives the collection of signals consisting of the original signal and N GCK-Fused signals , and each neural network outputs N + 1 classifications of the input vectors. The outputs are then combined based on majority vote.
Pre-process the data;
Create the clustered-knowledge database using RegPSO;
Separate the In-group fall data and ADL data into training and validation sets with ratio of 4:1;
Train MLP using resilient back propagation;
Create an RBF layer with cluster centroids taken from the clustered-knowledge result from RegPSO;
Pass the pre-processed data to the RBF layer;
Pass the output of the RBF layer to MLP layer;
Train the MLP with resilient back propagation;
Merge the RBF layer with the MLP layer.
In-group fall data
Collected from 5 healthy volunteers, 2 females and 3 males. Volunteers aged between 19 and 28 years.
293 fall signals were collected. Of these, 153 signals were used for training, and 140 signals used for testing (in-group performance)
Out-group fall data
Collected from 3 different healthy male volunteers whose data was not included in the training data. Volunteers aged between 19 and 28 years.
This set included 85 signals, all used to test "out-group" performance. The term "out-group" is used to indicated that these people's data was not used as training data.
The Activities of Daily Living (ADL) training data
Collected from 3 people. A total of 8 hours of ADL data was collected in a home environment. An additional hour of exercise data was recorded from 2 people in a gym environment. Volunteers aged between 19 and 28 years.
1831 ADL signals were collected. 1000 randomly selected ADL signals were used for the training set while 831 were used for testing. Of the 1000 randomly selected signals used for training, 750 related to ADL routine, and 250 related to ADL exercise. Of the 831 signals used for testing, 400 related to ADL routine and 381 related to ADL gym exercise.
Taken randomly from the training set with the ratio of training versus validation = 4:1.
Comparison of the results for the three classifier method
Ingroup Fall Sensitivity
(N = 140)
Outgroup Fall Sensitivity
(N = 85)
Routine ADL Specificity
(N = 450)
Exercise ADL Specificity
(N = 381)
Ensemble MLP + ARBF
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.
Limitations applicable to these results include the following. First, the number of subjects is relatively small. Second, the method for acquiring the falls data did not include truly accidental falls - the falls data was acquired when subjects deliberately fell onto a mattress. Third, the ADL data included only activities around the home and activities in the gym - data relating to ordinary work, transportation and other non-home activities would likely be more representative of some people's typical daily activities. Fourth, the data was acquired from people aged 19-28 years - it would be preferable for future work to include people from older age brackets.
These results are an exploratory step towards gaining knowledge about potential elements of a fall detection system. The implications of the results are somewhat limited due to limitations of the data acquisition processes. Notwithstanding, the results suggest the methods described here warrant further development and experimental investigation.
A further implication of these results is that some of the methods described here may also be applicable for body movement analysis and gait analysis relating to conditions that affect balance such as Parkinson's disease.
Future research will include acquiring data from different age groups, and developing methods to make use of data from ambulatory devices that include gyroscopes and a magnetometer.
Preliminary testing suggests the methods described here are noteworthy particularly for researchers who are attempting to improve the performance of ambulatory fall-detection systems. The methods should also be of interest for researchers who use (or are considering using) accelerometers to measure body movement. The 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 that the approach used here performs better than MLP alone.
The findings and conclusions in this report are those of the authors and do not necessarily represent the views of University of Technology Sydney or CSIRO.
- World Health Organization: WHO Global Report on Falls Prevention in Older Age. France 2007., ch.1: Google Scholar
- Barroso MC, Esteves GP, Nunes TP, MG Silva L, CD Faria A, Melo PL: A telemedicine instrument for remote evaluation of tremor: design and initial applications in fatigue and patients with Parkinson's Disease. BioMedical Engineering OnLine 2011., 10(14): doi:10.1186/1475–925X-10–14Google Scholar
- Moore ST, Dilda V, Hakim B, MacDougall HG: Validation of 24-hour ambulatory gait assessment in Parkinson's disease with simultaneous video observation. BioMedical Engineering OnLine 2011., 10(82): doi:10.1186/1475–925X-10–82Google Scholar
- Kangas M, Konttila A, Winblad I, Jämsä T: Determination of simple thresholds for accelerometry-based parameters for fall detection. Proc of the 29th Annual International Conference of the IEEE EMBS, Lyon 2007, 1367–1370.Google Scholar
- Chen J, Kwong K, Chang D, Luk J, Bajcsy R: Wearable Sensors for Reliable Fall Detection. Proc of the 27th Annual Conference of the IEEE EMBS, Shanghai 2005, 3551–3554.Google Scholar
- Bourke AK, VD Ven P, Gamble M, O'Connor R, Murphy K, Bogan E, McQuade E, Finucane P, ÓLaighin G, Nelson J: Assessment of Waist-worn Tri-Axial Accelerometer Based Fall-detection Algorithm using Continuous Unsupervised Activities. Proc of the 32nd Annual Conference of the IEEE EMBS, Buenos Aires 2010, 2782–2785.Google Scholar
- Shi G, Chan CS, Li WJ, Leung KS, Zou Y, Jin Y: Mobile Human Airbag System for Fall Protection Using MEMS Sensor and Embedded SVM Classifier. IEEE Sensors Journal 2009, 9(5):495–503.View ArticleGoogle Scholar
- Su SW, Chen W, Liu D, Kuang W, Yu X, Guo T, Celler BG, Nguyen HT: Dynamic Modelling of Heart Rate Response Under Different Exercise Intensity. The Open Medical Informatics Journal 2011, 4: 81–85.View ArticleGoogle Scholar
- Moulton B, Croucher G, Varis A, Chen J: Method for increasing the energy efficiency of wirelessly networked ambulatory health monitoring devices. Journal of Convergence Information Technology 2009, 5(1):7–14.Google Scholar
- Moulton B, Chaczko Z, Karatovic M: Data fusion and aggregation methods for pre-processing ambulatory monitoring and remote sensor data for upload to personal electronic health records. International Journal of Digital Content Technology and its Applications 2009, 3(4):120–127.Google Scholar
- Moulton B, Chaczko Z, Karatovic M: Updating electronic health records with information from sensor systems: considerations relating to standards and architecture arising from the development of a prototype system. Journal of Convergence Information Technology 2009, 4(4):21–26. 10.4156/jcit.vol4.issue4.3View ArticleGoogle Scholar
- Yuwono M, Handojoseno AMA, Nguyen HT: Optimization of Head Movement Recognition Using Augmented Radial Basis Function Neural Network. Proc of the 33rd Annual International Conference of the IEEE EMBS, Boston 2011, 2776–2779.Google Scholar
- Lajšner P, Kozub R: Using the MMA7360L ZSTAR2 Demo Board. Freescale Semiconductor 2007.Google Scholar
- Rioul O, Duhamel P: Fast Algorithm for Discrete and Continuous Wavelet Transforms. IEEE trans. on Information Theory 1992, 38(2):569–586. 10.1109/18.119724MathSciNetView ArticleMATHGoogle Scholar
- Merwe DW, Engelhrecht AP: Data Clustering using Particle Swarm Optimization. Congress on Evolutionary Computation 2003, 1: 215–220.Google Scholar
- Eberhart R, Kennedy J: A New Optimizer Using Particle Swarm Theory. IEEE Sixth International Symposium on Micro Machine and Human Science 1995, 39–43.View ArticleGoogle Scholar
- Evers GI, Ghalia MB: Regrouping Particle Swarm Optimization: A New Global Optimization Algorithm with Improved Performance Consistency Across Benchmarks. International Conference on Systems, Man, and Cybernetics 2009, San Antonio, TX, USA 2009, 3901–3908.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.