Energy Efficient Execution of POMDP Policies.

Recent advances in planning techniques for partially observable Markov decision processes (POMDPs) have focused on online search techniques and offline point-based value iteration. While these techniques allow practitioners to obtain policies for fairly large problems, they assume that a nonnegligible amount of computation can be done between each decision point. In contrast, the recent proliferation of mobile and embedded devices has lead to a surge of applications that could benefit from state-of-the-art planning techniques if they can operate under severe constraints on computational resources. To that effect, we describe two techniques to compile policies into controllers that can be executed by a mere table lookup at each decision point. The first approach compiles policies induced by a set of alpha vectors (such as those obtained by point-based techniques) into approximately equivalent controllers, while the second approach performs a simulation to compile arbitrary policies into approximately equivalent controllers. We also describe an approach to compress controllers by removing redundant and dominated nodes, often yielding smaller and yet better controllers. Further compression and higher value can sometimes be obtained by considering stochastic controllers. The compilation and compression techniques are demonstrated on benchmark problems as well as a mobile application to help persons with Alzheimer's to way-find. The battery consumption of several POMDP policies is compared against finite-state controllers learned using methods introduced in this paper. Experiments performed on the Nexus 4 phone show that finite-state controllers are the least battery consuming POMDP policies.

Upper limb contributions to frontal plane balance control in rollator-assisted walking.

While assisting with balance is a primary reason for rollator use, few studies have examined how the upper limbs are used for balance. This study examines upper limb contributions to balance control during rollator-assisted walking. We hypothesized that there would be an increased upper limb contribution, measured by mean vertical loading (Fz) and variation in frontal plane center-of-pressure (COPhigh), when walking balance is challenged/impaired. Experiment 1 compared straight-line and beam-walking in young adults (n = 11). As hypothesized, Fz and COPhighincreased in beam-walking compared to baseline (mean Fz: 13.7 vs. 9.1% body weight (BW), p < 0.001, RMS COPhigh: 1.35 vs. 1.07 cm, p < 0.001). Experiment 2 compared older adults who regularly use rollators (RU, n = 10) to older adult controls (CTL, n = 10). The predicted higher upper limb contribution in the RU group was not supported. However, when individuals were grouped by balance impairment, those with the lowest Berg Balance scores (< 45) demonstrated greater speed-adjusted COPhigh than those with higher scores (p = 0.013). Furthermore, greater COPhigh and Fz were correlated to greater reduction in step width, supporting the role of upper limb contributions to frontal plane balance. This work will guide studies assessing reliance on rollators by providing a basis for measurement of upper limb balance contributions.

Transparent muscle characterization using quantitative electromyography: different binarization mappings.

Evaluation of patients with suspected neuromuscular disorders is typically based on qualitative visual and auditory assessment of needle detected eletromyographic (EMG) signals; the resulting muscle characterization is subjective and highly dependent on the skill and experience of the examiner. Quantitative electromyography (QEMG) techniques were developed to extract motor unit potential trains (MUPTs) from needle detected EMG signals, and estimate features capturing motor unit potential (MUP) morphology and quantifying morphological consistency across MUPs belonging to the same MUPT. The aim of this study is to improve available methods for obtaining transparent muscle characterizations from features obtained using QEMG techniques. More specifically, we investigate the following. 1) Can the use of binarization mappings improve muscle categorization accuracies of transparent methods? 2) What are the appropriate binarization mappings in terms of accuracy and transparency? Results from four different sets of examined limb muscles (342 muscles in total) demonstrate that four out of the 10 investigated binarization mappings based on transparent characterization methods outperformed the multi-class characterizers based on Gaussian mixture models (GMM) and the corresponding binarization mappings based on GMM. This suggests that the use of an appropriate binarization mapping can overcome the decrease in categorization accuracy associated with quantizing MUPT features, which is necessary to obtain transparent characterizations. This performance gain can be attributed to the use of more relevant features and tuned quantization to obtain more specific binary characterizations.

Measuring life space in older adults with mild-to-moderate Alzheimer's disease using mobile phone GPS.

BACKGROUND: As an indicator of physical and cognitive functioning in community-dwelling older adults, there is increasing interest in measuring life space, defined as the geographical area a person covers in daily life. Typically measured through questionnaires, life space can be challenging to assess in amnestic dementia associated with Alzheimer's disease (AD). While global positioning system (GPS) technology has been suggested as a potential solution, there remains a lack of data validating GPS-based methods to measure life space in cognitively impaired populations. OBJECTIVE: The purpose of the study was to evaluate the construct validity of a GPS system to provide quantitative measurements of global movement for individuals with mild-to-moderate AD. METHODS: Nineteen community-dwelling older adults with mild-to-moderate AD (Mini-Mental State Examination score 14-28, age 70.7 ± 2.2 years) and 33 controls (CTL; age 74.0 ± 1.2 years) wore a GPS-enabled mobile phone during the day for 3 days. Measures of geographical territory (area, perimeter, mean distance from home, and time away from home) were calculated from the GPS log. Following a log-transformation to produce symmetrical distributions, group differences were tested using two-sample t tests. Construct validity of the GPS measures was tested by examining the correlation between the GPS measures and indicators of physical function [steps/day, gait velocity, and Disability Assessment for Dementia (DAD)] and affective state (Apathy Evaluation Scale and Geriatric Depression Scale). Multivariate regression was performed to evaluate the relative strength of significantly correlated factors. RESULTS: GPS-derived area (p < 0.01), perimeter (p < 0.01), and mean distance from home (p < 0.05) were smaller in the AD group compared to CTL. The correlation analysis found significant associations of the GPS measures area and perimeter with all measures of physical function (steps/day, DAD, and gait velocity; p < 0.01), symptoms of apathy (p < 0.01), and depression (p < 0.05). Multivariate regression analysis indicated that gait velocity and dependence were the strongest variables associated with GPS measures. CONCLUSION: This study demonstrated that GPS-derived area and perimeter: (1) distinguished mild-to-moderate AD patients from CTL and (2) were strongly correlated with physical function and affective state. These findings confirm the ability of GPS technology to assess life space behaviour and may be particularly valuable to continuously monitor functional decline associated with neurodegenerative disease, such as AD.

Towards a mobility diagnostic tool: tracking rollator users' leg pose with a monocular vision system.

Cognitive assistance of a rollator (wheeled walker) user tends to reduce the attentional capacity of the user and may impact her stability. Hence, it is important to understand and track the pose of rollator users before augmenting a rollator with some form of cognitive assistance. While the majority of current markerless vision systems focus on estimating 2D and 3D walking motion in the sagittal plane, we wish to estimate the 3D pose of rollator users' lower limbs from observing image sequences in the coronal (frontal) plane. Our apparatus poses a unique set of challenges: a single monocular view of only the lower limbs and a frontal perspective of the rollator user. Since motion in the coronal plane is relatively subtle, we explore multiple cues within a Bayesian probabilistic framework to formulate a posterior estimate for a given subject's leg limbs. In this work, our focus is on evaluating the appearance model (the cues). Preliminary experiments indicate that texture and colour cues conditioned on the appearance of a rollator user outperform more general cues, at the cost of manually initializing the appearance offline.

A planning system based on Markov decision processes to guide people with dementia through activities of daily living.

Older adults with dementia often cannot remember how to complete activities of daily living and require a caregiver to aid them through the steps involved. The use of a computerized guidance system could potentially reduce the reliance on a caregiver. This paper examines the design and preliminary evaluation of a planning system that uses Markov decision processes (MDPs) to determine when and how to provide prompts to a user with dementia for guidance through the activity of handwashing. Results from the study suggest that MDPs can be applied effectively to this type of guidance problem. Considerations for the development of future guidance systems are presented.