Shuang Cang

Elderly activities recognition and classification for applications in assisted living

Assisted living systems can help support elderly persons with their daily activities in order to help them maintain healthy and safety while living independently. However, most current systems are ineffective in actual situation, difficult to use and have a low acceptance rate. There is a need for an assisted living solution to become intelligent and also practical issues such as user acceptance and usability need to be resolved in order to truly assist elderly people. Small, inexpensive and low-powered consumption sensors are now available which can be used in assisted living applications to provide sensitive and responsive services based on users current environments and situations. This paper aims to address the issue of how to develop an activity recognition method for a practical assisted living system in term of user acceptance, privacy (non-visual) and cost. The paper proposes an activity recognition and classification method for detection of Activities of Daily Livings (ADLs) of an elderly person using small, low-cost, non-intrusive non-stigmatize wrist worn sensors. Experimental results demonstrate that the proposed method can achieve a high classification rate (>90%). Statistical tests are employed to support this high classification rate of the proposed method. Also, we prove that by combining data from temperature sensor and/or altimeter with accelerometer, classification accuracy can be improved.

A practical multi-sensor activity recognition system for home-based care

To cope with the increasing number of aging population, a type of care which can help prevent or postpone entry into institutional care is preferable. Activity recognition can be used for home-based care in order to help elderly people to remain at home as long as possible. This paper proposes a practical multi-sensor activity recognition system for home-based care utilizing on-body sensors. Seven types of sensors are investigated on their contributions toward activity classification. We collected a real data set through the experiments participated by a group of elderly people. Seven classification models are developed to explore contribution of each sensor. We conduct a comparison study of four feature selection techniques using the developed models and the collected data. The experimental results show our proposed system is superior to previous works achieving 97% accuracy. The study also demonstrates how the developed activity recognition model can be applied to promote a home-based care and enhance decision support system in health care. Propose a practical multi-sensor activity recognition system for home-based care.Collect a real data set from a group of elderly people using seven on-body sensors.Conduct investigation on the effect of different sensor in human activity classification.Evaluate different feature selection and classification techniques for activity recognition.

A combination selection algorithm on forecasting

It is widely accepted in forecasting that a combination model can improve forecasting accuracy. One important challenge is how to select the optimal subset of individual models from all available models without having to try all possible combinations of these models. This paper proposes an optimal subset selection algorithm from all individual models using information theory. The experimental results in tourism demand forecasting demonstrate that the combination of the individual models from the selected optimal subset significantly outperforms the combination of all available individual models. The proposed optimal subset selection algorithm provides a theoretical approach rather than experimental assessments which dominate literature.

Mutual information based input feature selection for classification problems

The elimination process aims to reduce the size of the input feature set and at the same time to retain the class discriminatory information for classification problems. This paper investigates the approaches to solve classification problems of the feature selection and proposes a new feature selection algorithm using the mutual information (MI) concept in information theory for the classification problems. The proposed algorithm calculates the MI between the combinations of input features and the class instead of the MI between a single input feature and the class for both continuous-valued and discrete-valued features. Three experimental tests are conducted to evaluate the proposed algorithm. Comparison studies of the proposed algorithm with the previously published classification algorithms indicate that the proposed algorithm is robust, stable and efficient.

Genetic Algorithm-Based Classifiers Fusion for Multisensor Activity Recognition of Elderly People

Activity recognition of an elderly person can be used to provide information and intelligent services to health care professionals, carers, elderly people, and their families so that the elderly people can remain at homes independently. This study investigates the use and contribution of wrist-worn multisensors for activity recognition. We found that accelerometers are the most important sensors and heart rate data can be used to boost classification of activities with diverse heart rates. We propose a genetic algorithm-based fusion weight selection (GAFW) approach which utilizes GA to find fusion weights. For all possible classifier combinations and fusion methods, the study shows that 98% of times GAFW can achieve equal or higher accuracy than the best classifier within the group.

A self-adaptive image normalization and quaternion PCA based color image watermarking algorithm

This paper proposes a novel robust digital color image watermarking algorithm which combines color image feature point extraction, shape image normalization and QPCA (quaternion principal component algorithm) based watermarking embedding (QWEMS) and extraction (QWEXS) schemes. The feature point extraction method called Mexican Hat wavelet scale interaction is used to select the points which can survive various attacks and also be used as reference points for both watermarking embedding and extraction. The normalization shape image of the local quadrangle image of which the four corners are feature points of the original image is invariant to translation, rotation, scaling and skew, by which we can obtain the relationship between the feature images of the original image and the watermarked image which has suffered with geometrical attacks. The proposed QWEMS and QWEXS schemes which denote the color pixel as a pure quaternion and the feature image as a quaternion matrix can improve the robustness and the imperceptibility of the embedding watermarking. To simplify the eigen-decomposition procedure of the quaternion matrix, we develop a calculation approach with which the eigen-values and the corresponding eigen-vectors of the quaternion matrix can be computed. A binary watermark image is embedded in the principal component coefficients of the feature image. Simulation results demonstrate that the proposed algorithm can survive a variety of geometry attacks, i.e. translation, rotation, scaling and skew, and can also resist the attacks of many signal processing procedures, for example, moderate JPEG compression, salt and pepper noise, Gaussian filtering, median filtering, and so on.

Delay, Reliability, and Throughput Based QoS Profile: A MAC Layer Performance Optimization Mechanism for Biomedical Applications in Wireless Body Area Sensor Networks

Recently, increasing demand for remote healthcare monitoring systems poses a specific set of Quality of Services (QoS) requirements to the MAC layer protocols and standards (IEEE 802.15.6, IEEE 802.15.4, etc.) of Wireless Body Area Sensor Networks (WBASNs). They mainly include time bounded services (latency), reliable data transmission, fair channel distribution, and specified data rates. The existing MAC protocols of WBASNs are lack of a specific set of QoS. To address this, the paper proposes a QoS profile named delay, reliability, and throughput (DRT). The QoS values computed through DRT profile provide maximum reliability of data transmission within an acceptable latency and data rates. The DRT is based on the carrier sense multiple access with collision avoidance (CSMA/CA) channel access mechanism and considers IEEE 802.15.4 (low-rate WPAN) and IEEE 802.15.6 (WBASN). Further, a detailed performance analysis of different frequency bands is done which are standardized for WBASNs, that is, 420 MHz, 868 MHz, 2.4 GHz, and so forth. Finally, a series of experiments are conducted to produce statistical results for DRT profile with respect to delay, reliability, and packet delivery ratio (PDR). The calculated results are verified through extensive simulations in the CASTALIA 3.2 framework using the OMNET++ network simulator.

Impact of Load Variation on Joint Angle Estimation from Surface EMG Signals

Many studies use surface electromyogram (sEMG) signals to estimate the joint angle, for control of upper-limb exoskeletons and prostheses. However, several practical factors still affect its clinical applicability. One of these factors is the load variation during daily use. This paper demonstrates that the load variation can have a substantial impact on performance of elbow angle estimation. This impact leads an increase in mean RMSE (Root-Mean-Square Error) from 7.86 ° to 20.44 ° in our experimental test. Therefore, we propose three methods to address this issue: 1) pooling the training data from all loads together to form the pooled training data for the training model; 2) adding the measured load value (force sensor) as an additional input; and 3) developing a two-step hybrid estimation approach based on load and sEMG. Experiments are conducted with five subjects to investigate the feasibility of the proposed three methods. The results show that the mean RMSE is reduced from 20.44 ° to 13.54 ° using method one, 10.47 ° using method two, and 8.48 ° using method three, respectively. Our study indicates that 1) the proposed methods can improve performance and stability on joint angle estimation and 2) sensor fusion (sEMG sensor and force sensor) is an efficient way to resolve the adverse effect of load variation.

Assistive technology for people with dementia: an overview and bibliometric study

BACKGROUND This study presents an overview of recent research activities in assistive technology (AT) for people with dementia. Bibliometric studies are used to explore breadth and depth of different research areas, yet this method has not yet been fully utilised in AT research for people with dementia. METHODS The bibliometric method was used for collecting studies related to AT. Based on inclusion/exclusion criteria, the AT studies with a focus on people with dementia are considered. STUDY SCOPE The study is based on factors such as number of publications, citations per paper, collaborative research output, P-Index, major research and application areas and national dementia strategies. DATA COLLECTION Data were collected from 2000 to 2014 in AT research. The top 10 countries are selected based on their research outputs. RESULTS USA emerged as the leading contributor with 503 publications and an annual growth rate of 16%, followed by UK with 399 publications and growth rate of 22%. Germany with 101 publications is on the 6th place, but it has a higher citation rate 16.43% as compared to USA (13.34%). Although all 10 countries show good collaborative research output, Italy, Spain and the Netherlands emerge as top collaborative research contributors with high percentages (84%, 84% and 79%). All the top 10 countries, except Canada, Germany and Spain, have national dementia strategies in place. CONCLUSION The overall analysis shows that USA and UK are working extensively in AT research for people with dementia. Both these countries also have well established national dementia strategies.

Maximum relevancy maximum complementary feature selection for multi-sensor activity recognition

We propose a feature selection algorithm using MRMC.Show that MRMC provides a good result comparing to the 3 popular algorithms.The complementary measure improves the performance of the Clamping algorithm.Evaluate the proposed algorithm on 2 well-defined problems and 5 real life data sets. In the multi-sensor activity recognition domain, the input space is often large and contains irrelevant and overlapped features. It is important to perform feature selection in order to select the smallest number of features which can describe the outputs. This paper proposes a new feature selection algorithms using the maximal relevance and maximal complementary (MRMC) based on neural networks. Unlike other feature selection algorithms that are based on relevance and redundancy measurements, the idea of how a feature complements to the already selected features is utilized. The proposed algorithm is evaluated on two well-defined problems and five real world data sets. The data sets cover different types of data i.e. real, integer and category and sizes i.e. small to large set of features. The experimental results show that the MRMC can select a smaller number of features while achieving good results. The proposed algorithm can be applied to any type of data, and demonstrate great potential for the data set with a large number of features.