Sawsan M. Mahmoud

Smart homes for the elderly dementia sufferers: identification and prediction of abnormal behaviour

In this paper, we have described a solution for supporting independent living of the elderly by means of equipping their home with a simple sensor network to monitor their behaviour. Standard home automation sensors including movement sensors and door entry point sensors are used. By monitoring the sensor data, important information regarding any anomalous behaviour will be identified. Different ways of visualizing large sensor data sets and representing them in a format suitable for clustering the abnormalities are also investigated. In the latter part of this paper, recurrent neural networks are used to predict the future values of the activities for each sensor. The predicted values are used to inform the caregiver in case anomalous behaviour is predicted in the near future. Data collection, classification and prediction are investigated in real home environments with elderly occupants suffering from dementia.

Behavioural pattern identification and prediction in intelligent environments

In this paper, the application of soft computing techniques in prediction of an occupants behaviour in an inhabited intelligent environment is addressed. In this research, daily activities of elderly people who live in their own homes suffering from dementia are studied. Occupancy sensors are used to extract the movement patterns of the occupant. The occupancy data is then converted into temporal sequences of activities which are eventually used to predict the occupant behaviour. To build the prediction model, different dynamic recurrent neural networks are investigated. Recurrent neural networks have shown a great ability in finding the temporal relationships of input patterns. The experimental results show that non-linear autoregressive network with exogenous inputs model correctly extracts the long term prediction patterns of the occupant and outperformed the Elman network. The results presented here are validated using data generated from a simulator and real environments.

Abnormal behaviours identification for an elder's life activities using dissimilarity measurements

Identifying abnormal behaviour is an important factor in activity recognition. The aim of this paper is to design a system able to detect the abnormal behaviours of daily activity living in an intelligent environment. We approach this by applying dissimilarity (distance) measures on data collected from a single inhabitant environment. The data are acquired from occupancy sensors such as a door and motion sensors. Since the data is collected from these sensors has a discrete value either on or off, only the binary dissimilarity measures are considered in this paper. There are several distance measurements which find the mismatching bits of two binary data sets. In this paper, two major dissimilarity measures, which include hamming distance and fuzzy hamming distance, are used and compared. These measures can help in distinguishing between normal and abnormal behaviour patterns in order to improve the quality of elderly peoples lives. Two case studies where the inhabitants suffer from dementia are used to verify the accuracy of the results. The experimental results demonstrate that fuzzy hamming distance gives a smaller distance than classic hamming distance in the case of motion sensors over door sensors.

User Activities Outliers Detection; Integration of Statistical and Computational Intelligence Techniques

In this article, a hybrid technique for user activities outliers detection is introduced. The hybrid technique consists of a two‐stage integration of principal component analysis and fuzzy rule‐based systems. In the first stage, the Hamming distance is used to measure the differences between different activities. Principal component analysis is then applied to the distance measures to find two indices of Hotellings T2 and squared prediction error. In the second stage of the process, the calculated indices are provided as inputs to the fuzzy rule‐based systems to model them heuristically. The model is used to identify the outliers and classify them. The proposed system is tested in real home environments, equipped with appropriate sensory devices, to identify outliers in the activities of daily living of the user. Three case studies are reported to demonstrate the effectiveness of the proposed system. The proposed system successfully identifies the outliers in activities distinguishing between the normal and abnormal behavioral patterns.

Trend Modelling of Elderly Lifestyle within an Occupancy Simulator

In this paper, the trend as an important component in activities of daily living is modelled and integrated to a single-occupant occupancy simulator. Therefore, in the occupancy signal generated by the simulator, both seasonality and trend are included in occupants movements. As the result of trends integrated to the simulator, occupancy signals with different types of trends such as increasing, decreasing, cyclic and chaotic trends can be generated. Different types of the trend in occupancy signal improves the occupancy modelling by enabling the model to incorporate long term differentiation in occupants behaviour i.e. ageing, health, and other changes in his/her activities of daily living. In this paper, the effect of trends in the occupancy signal generated by the modified simulator is tested by applying autocorrelation function.

User activities outlier detection system using principal component analysis and fuzzy rule-based system

In this paper, a user activities outlier detection system is introduced. The proposed system is implemented in a smart home environment equipped with appropriate sensory devices. An activity outlier detection system consist of a two-stage integration of Principal Component Analysis (PCA) and Fuzzy Rule-Based System (FRBS). In the first stage, the Hamming distance is used to measure the distances between the activities. PCA is then applied to the distance measures to find two indices of Hotellings T2 and Squared Prediction Error (SPE). In the second stage of the process, the calculated indices are provided as inputs to FRBSs to model them heuristically. They are used to identify the outliers and classify them. Three case studies are reported to demonstrate the effectiveness of the proposed system. The proposed system successfully identifies the outliers and helps in distinguishing between the normal and abnormal behaviour patterns of the Activities of Daily Living (ADL).