Louis Atallah

Sensor Positioning for Activity Recognition Using Wearable Accelerometers

Activities of daily living are important for assessing changes in physical and behavioral profiles of the general population over time, particularly for the elderly and patients with chronic diseases. Although accelerometers have been used widely in wearable devices for activity classification, the positioning of the sensors and the selection of relevant features for different activity groups still pose significant research challenges. This paper investigates wearable sensor placement at different body positions and aims to provide a systematic framework that can answer the following questions: 1) What is the ideal sensor location for a given group of activities? and 2) Of the different time-frequency features that can be extracted from wearable accelerometers, which ones are the most relevant for discriminating different activity types?

Review: The use of pervasive sensing for behaviour profiling - a survey

With the maturity of sensing and pervasive computing techniques, extensive research is being carried out in using different sensing techniques for understanding human behaviour. An introduction to key modalities of pervasive sensing is presented. Behaviour modelling is then highlighted with a focus on probabilistic models. The survey discusses discriminative approaches as well as relevant work on behaviour pattern clustering and variability. The influence of interacting with people and objects in the environment is also discussed. Finally, challenges and new research opportunities are highlighted.

Sensor Placement for Activity Detection Using Wearable Accelerometers

Activities of daily living are important for assessing changes in physical and behavioural profiles of the general population over time, particularly for the elderly and patients with chronic diseases. Although accelerometers are widely integrated with wearable sensors for activity classification, the positioning of the sensors and the selection of relevant features for different activity groups still pose interesting research challenges. This paper investigates wearable sensor placement at different body positions and aims to provide a framework that can answer the following questions: (i) What is the ideal sensor location for a given group of activities? (ii) Of the different time-frequency features that can be extracted from wearable accelerometers, which ones are most relevant for discriminating different activity types?

Direction sensitive fall detection using a triaxial accelerometer and a barometric pressure sensor

Falling is one of the leading causes of serious health decline or injury-related deaths in the elderly. For survivors of a fall, the resulting health expenses can be a devastating burden, largely because of the long recovery time and potential comorbidities that ensue. The detection of a fall is, therefore, important in care of the elderly for decreasing the reaction time by the care-givers especially for those in care who are particularly frail or living alone. Recent advances in motion-sensor technology have enabled wearable sensors to be used efficiently for pervasive care of the elderly. In addition to fall detection, it is also important to determine the direction of a fall, which could help in the location of joint weakness or post-fall fracture. This work uses a waist-worn sensor, encompassing a 3D accelerometer and a barometric pressure sensor, for reliable fall detection and the determination of the direction of a fall. Also assessed is an efficient analysis framework suitable for on-node implementation using a low-power micro-controller that involves both feature extraction and fall detection. A detailed laboratory analysis is presented validating the practical application of the system.

Real-Time Activity Classification Using Ambient and Wearable Sensors

New approaches to chronic disease management within a home or community setting offer patients the prospect of more individually focused care and improved quality of life. This paper investigates the use of a light-weight ear worn activity recognition device combined with wireless ambient sensors for identifying common activities of daily living. A two-stage Bayesian classifier that uses information from both types of sensors is presented. Detailed experimental validation is provided for datasets collected in a laboratory setting as well as in a home environment. Issues concerning the effective use of the relatively limited discriminative power of the ambient sensors are discussed. The proposed framework bodes well for a multi-dwelling environment, and offers a pervasive sensing environment for both patients and care-takers.

A Pervasive Body Sensor Network for Measuring Postoperative Recovery at Home

Patients going home following major surgery are susceptible to complications such as wound infection, abscess formation, malnutrition, poor analgesia, and depression, all of which can develop after the fifth postoperative day and slow recovery. Although current hospital recovery monitoring systems are effective during perioperative and early postoperative periods, they cannot be used when the patient is at home. Measuring and quantifying home recovery is currently a subjective and labor-intensive process. This case report highlights the development and piloting of a wireless body sensor network to monitor postoperative recovery at home in patients undergoing abdominal surgery. The device consists of wearable sensors (vital signs, motion) combined with miniaturized computers wirelessly linked to each other, thus allowing continuous monitoring of patients in a pervasive (unobtrusive) manner in any environment. Initial pilot work with results in both the simulated (with volunteers) and the real home environment (with patients) is presented.

Changes in prefrontal cortical behaviour depend upon familiarity on a bimanual co-ordination task: an fNIRS study.

To investigate neurocognitive mechanisms associated with task-related expertise development, this paper investigates serial changes in prefrontal activation patterns using functional near infrared spectroscopy (fNIRS). We evaluate cortical function in 62 healthy subjects with varying experience during serial evaluations of a knot-tying task. All tasks were performed bimanually and self paced, with fixed episodes of motor rest for five repetitions. Improvements in technical skill were evaluated using dexterity indices to quantify time, total movements and pathlength required to complete trials. Significant improvements in technical skills were observed in novices between the 2nd and 3rd trials, associated with increasing task familiarity. In trained subjects, minimal fluctuation in task-related oxyhaemoglobin (HbO(2)) and deoxyhaemoglobin (HHb) changes were observed in association with more stable task performance. In contrast, two significant transitions in prefrontal haemodynamic change were observed in novices. Greater task-related increases in HbO(2) and decreases in HHb were identified on the second trial compared to the first. Relative decreases in HbO(2) and increases in HHb change were observed between the third and fourth, and fourth and fifth trials respectively. These data suggest that prefrontal processing across five knot-tying trials is influenced by the level of experience on a task. Modifications in prefrontal activation appear to confer technical performance adaptation in novices.

HMM assessment of quality of movement trajectory in laparoscopic surgery

Laparoscopic surgery poses many different constraints for the operating surgeon, resulting in a slow uptake of advanced laparoscopic procedures. Traditional approaches to the assessment of surgical performance rely on prior classification of a cohort of surgeons’ technical skills for validation, which may introduce subjective bias to the outcome. In this study, Hidden Markov Models (HMMs) are used to learn surgical maneuvers from 11 subjects with mixed abilities. By using the leave-one-out method, the HMMs are trained without prior clustering of subjects into different skill levels, and the output likelihood indicates the similarity of a particular subjects motion trajectories to those of the group. The results show that after a short period of training, the novices become more similar to the group when compared to the initial pre-training assessment. The study demonstrates the strength of the proposed method in ranking the quality of trajectories of the subjects, highlighting its value in minimizing the subjective bias in skills assessment for minimally invasive surgery.

Real-Time Pervasive Monitoring for Postoperative Care

Post surgical care is an important part of the surgical recovery process. With the introduction of minimally invasive surgery (MIS), the recovery time of patients has been shortened significantly. This has led to a shift of postoperative care from hospital to home environment. To prevent the occurrence of adverse events, the care of these patients is mainly relied on routine visits by home-care nurses. This type of episodic examination can only capture a snapshot of the overall recovery process, and many early signs of potential complication can go undetected. The development of Body Sensor Networks (BSNs) has enabled the use of miniaturised wireless sensors for continuous monitoring of postoperative patients. This paper examines the potential of processing-on-node algorithms for further reducing the wireless bandwidth, and therefore the overall power consumption of the sensors. The accuracy and robustness of the technique are demonstrated with lab experiments and a preliminary clinical case study.

An Intelligent Food-Intake Monitoring System Using Wearable Sensors

The prevalence of obesity worldwide presents a great challenge to existing healthcare systems. There is a general need for pervasive monitoring of the dietary behaviour of those who are at risk of co-morbidities. Currently, however, there is no accurate method of assessing the nutritional intake of people in their home environment. Traditional methods require subjects to manually respond to questionnaires for analysis, which is subjective, prone to errors, and difficult to ensure consistency and compliance. In this paper, we present a wearable sensor platform that autonomously provides detailed information regarding a subjects dietary habits. The sensor consists of a microphone and a camera and is worn discretely on the ear. Sound features are extracted in real-time and if a chewing activity is classified, the camera captures a video sequence for further analysis. From this sequence, a number of key frames are extracted to represent important episodes during the course of a meal. Results show a high classification rate of chewing activities, and the visual log demonstrates a detailed overview of the subjects food intake that is difficult to quantify from manually-acquired food records.