Modelling large-scale information systems using ADLs – An industrial experience report

An organisation that had developed a large information system wanted to embark on a programme that would involve large-scale evolution of it. As a precursor to this, it was decided to create a comprehensive architectural description to capture and understand the system’s design. This undertaking faced a number of challenges, including a low general awareness of software modelling and software architecture practices. The approach taken by the software architects tasked with this project included the definition of a simple, very specific, architecture description language. This paper reports our experience of the project and a simple ADL that we created as part of it. 

Measurement and correlates of caregiver self-efficacy amongst family caregivers of persons with dementia living in Shanghai, China

Background: Caring for family members with dementia can be a long-term, burdensome task resulting in physical and emotional distress and impairment. Research has demonstrated significantly lower levels of selfefficacy among family caregivers of people with dementia (CGs) than caregivers of relatives with non-dementia diseases. Intervention studies have also suggested that the mental and physical health of dementia CGs could be improved through the enhancement of their self-efficacy. However, studies are limited in terms of the influences of caregiver self-efficacy on caregiver behaviour, subjective burden and health-related quality of life. Of particular note is that there are no studies on the applicability of caregiver self-efficacy in the social context of China. Objective: The purpose of this thesis was to undertake theoretical exploration using Bandura’s (1997) self-efficacy theory to 1) revise the Revised Caregiving Self-Efficacy Scale (C-RCSES) (Steffen, McKibbin, Zeiss, Gallagher-Thompson, & Bandura, 2002), and 2) explore determinants of caregiver self-efficacy and the role of caregiver self-efficacy and other conceptual constructs (including CGs’ socio-demographic characteristics, CRs’ impairment and CGs’ social support) in explaining and predicting caregiver behaviour, subjective burden and health-related quality of life among CGs in China. Methodology: Two studies were undertaken: a qualitative elicitation study with 10 CGs; and a cross-sectional survey with 196 CGs. In the first study, semi-structured interviews were conducted to explore caregiver behaviours and corresponding challenges for their performance. The findings of the study assisted in the development of the initial items and domains of the Chinese version of the Revised Caregiving Self-Efficacy Scale (C-RCSES). Following changes to items in the scale, the second study, a cross-sectional survey with 196 CGs was conducted to evaluate the psychometric properties of C-RCSES and to test a hypothesised self-efficacy model of family caregiving adapted from Bandura’s theory (1997). Results: 35 items were generated from the qualitative data. The content validity of the C-RCSES was assessed and ensured in Study One before being used for the cross-sectional survey. Eight items were removed and five subscales (caregiver self-efficacy for gathering information about treatment, symptoms and health care; obtaining support; responding to problematic behaviours; management of household, personal and medical care; and controlling upsetting thoughts about caregiving) were identified after principal component factor analysis on the cross-sectional survey data. The reliability of the scale is acceptable: the Cronbach’s alpha coefficients for the whole scale and for each subscale were all over .80; and the fourweek test-retest reliabilities for the whole scale and for each subscale ranged from .64 to .85. The concurrent, convergent and divergent validity were also acceptable. CGs reported moderate levels of caregiver self-efficacy. Furthermore, the level of self-efficacy for management of household, personal and medical care was relatively high in comparison to those of the other four domains of caregiver self-efficacy. Caregiver self-efficacy was also significantly influenced by CGs’ socio-demographic characteristics and the caregiving external factors (CR impairment and social support that CGs obtained). The level of caregiver behaviour that CGs reported was higher than that reported in other Chinese research. CGs’ socio-demographics significantly influenced caregiver behaviour, whereas caregiver self-efficacy did not influence caregiver behaviour. Regarding the two external factors, CGs who cared for highly impaired relatives reported high levels of caregiver behaviour, but social support did not influence caregiver behaviour. Regarding caregiver subjective burden and health-related quality of life, CGs reported moderate levels of subjective burden, and their level of healthrelated quality of life was significantly lower than that of the general population in China. The findings also indicated that CGs’ subjective burden and health-related quality of life were influenced by all major factors in the hypothesised model, including CGs’ socio-demographics, CRs’ impairment, social support that CGs obtained, caregiver self-efficacy and caregiver behaviour. Of these factors, caregiver self-efficacy and social support significantly improved their subjective burden and health-related quality of life; whereas caregiver behaviour and CRs’ impairment were detrimental to CGs, such as increasing subjective burden and worsening health-related quality of life. Conclusion: While requiring further exploration, the qualitative study was the first qualitative research conducted in China to provide an in-depth understanding of CGs’ caregiving experience, including their major caregiver behaviours and the corresponding challenges. Meanwhile, although the C-RCSES needs further psychometric testing, it is a useful tool for assessing caregiver self-efficacy in Chinese populations. Results of the qualitative and quantitative study provide useful information for future studies regarding the explanatory power of caregiver self-efficacy to caregiver behaviour, subjective burden and health-related quality of life. Additionally, integrated with Bandura’s theory, the findings from the quantitative study also suggested a further study exploring the role of outcome expectations in caregiver behaviour, subjective burden and healthrelated quality of life.

The decline in activities of daily living at discharge (DADLD) index : stratifying patients at lower and higher risk

Background: Decreased ability to perform Activities of Daily Living (ADLs) during hospitalisation has negative consequences for patients and health service delivery. Objective: To develop an Index to stratify patients at lower and higher risk of a significant decline in ability to perform ADLs at discharge. Design: Prospective two cohort study comprising a derivation (n=389; mean age 82.3 years; SD� 7.1) and a validation cohort (n=153; mean age 81.5 years; SD� 6.1). Patients and setting: General medical patients aged = 70 years admitted to three university-affiliated acute care hospitals in Brisbane, Australia. Measurement and main results: The short ADL Scale was used to identify a significant decline in ability to perform ADLs from premorbid to discharge. In the derivation cohort, 77 patients (19.8%) experienced a significant decline. Four significant factors were identified for patients independent at baseline: 'requiring moderate assistance to being totally dependent on others with bathing'; 'difficulty understanding others (frequently or all the time)'; 'requiring moderate assistance to being totally dependent on others with performing housework'; a 'history of experiencing at least one fall in the previous 90 days prior to hospital admission' in addition to 'independent at baseline', which was protective against decline at discharge. 'Difficulty understanding others (frequently or all the time)' and 'requiring moderate assistance to being totally dependent on others with performing housework' were also predictors for patients dependent in ADLs at baseline. Sensitivity, specificity, Positive Predictive Value (PPV), and Negative Predictive Value (NPV) of the DADLD dichotomised risk scores were: 83.1% (95% CI 72.8; 90.7); 60.5% (95% CI 54.8; 65.9); 34.2% (95% CI 27.5; 41.5); 93.5% (95% CI 89.2; 96.5). In the validation cohort, 47 patients (30.7%) experienced a significant decline. Sensitivity, specificity, PPV and NPV of the DADLD were: 78.7% (95% CI 64.3; 89.3); 69.8% (95% CI 60.1, 78.3); 53.6% (95% CI 41.2; 65.7); 88.1% (95% CI 79.2; 94.1). Conclusions: The DADLD Index is a useful tool for identifying patients at higher risk of decline in ability to perform ADLs at discharge.

Towards more flexible Architecture Description Languages for industrial applications

Architecture Description Languages (ADLs) have emerged in recent years as a tool for providing high-level descriptions of software systems in terms of their architectural elements and the relationships among them. Most of the current ADLs exhibit limitations which prevent their widespread use in industrial applications. In this paper, we discuss these limitations and introduce ALI, an ADL that has been developed to address such limitations. The ALI language provides a rich and flexible syntax for describing component interfaces, architectural patterns, and meta-information. Multiple graphical architectural views can then be derived from ALI's textual notation.

Tele-Supervised FES-Assisted Exercise for Hemiplegic Upper Limb

Stroke survivors often have upper limb (UL) hemiparesis, limiting their ability to perform activities of daily life (ADLs). Intensive, task-oriented exercise therapy (ET) can improve UL function, but motivation to perform sufficient ET is difficult to maintain. Here we report on a trial in which a workstation was deployed in the homes of chronic stroke survivors to enable tele-coaching of ET in the guise of computer games. Participants performed 6 weeks of 1 hour/day, 5 days/week ET. Hand opening and grasp were assisted with functional electrical stimulation (FES). The primary outcome measure was the Action Research Arm Test (ARAT). Secondary outcome measures included a quantitative test of UL function performed on the workstation, grasp force measurements and transcranial magnetic stimulation (TMS). Improvements were seen in the functional tests, but surprisingly, not in the TMS responses. An important finding was that participants commencing with intermediate functional scores improved the most.

CONCLUSIONS: 1) Daily, tele-supervised FES-ET in chronic stroke survivors is feasible with commercially-available technology. 2) The intervention can significantly improve UL function, particularly in people who start with an intermediate level of function. 3) Significant improvements in UL function can occur in the absence of changes in TMS responses.

Moderators of the Relationship between Family Caregiver Proxy-Ratings and Person with Dementia Self-Ratings of Quality of Life

As the dementia spectrum lacks any viable cure, quality of life is typically regarded as an essential measure of assessing the clinical course and evaluating interventions. With caregivers typically providing this rating to health professionals, the literature has noted inconsistencies between caregiver and person with dementia (PwD) ratings of quality of life and suggested several factors may moderate the rating relationship. To investigate this, an intraclass correlation coefficient was calculated to observe rating agreement and moderator regression analysis was conducted to explore potential moderators. Potential moderators of caregiver burden, caregiver age, caregiver income, PwD IADLs/ADLs, PwD education, PwD cognitive impairment, PwD depressive symptom severity, PwD behavioural symptom severity, as well as relationship between caregiver and PwD. Utilizing secondary data from 107 recruited dyads, analyses conducted found fair agreement between caregivers and those with dementia while none of the hypothesized factors were found to moderate the rating relationship.

Tendances et déterminants de l’utilisation de l’aide à domicile subventionnée par le gouvernement par les personnes âgées au Canada, 1996-1997 à 2009-2010

Il est connu que le vieillissement de la population canadienne va augmenter la demande d’aide à domicile et qu’un soutien insuffisant du secteur public accroît l’implication des réseaux informels et privés. En s’appuyant sur les données de l’Enquête nationale sur la santé de la population (ENSP) et de l’Enquête sur la santé dans les collectivités canadiennes (ESCC), cette recherche analyse les tendances et déterminants de l’utilisation de l’aide à domicile subventionnée par le gouvernement au Canada et en Ontario entre 1996-1997 et 2009-2010. Il en ressort que la proportion de bénéficiaires âgés de 65 ans et plus est en diminution, passant de 9,9% à 9,2% au Canada (1996-1997 à 2005) et de 10,2% à 8,8% puis 9,6% en Ontario (1996-1997, 2005, 2009-2010). Les compressions sont particulièrement ressenties dans les services de maintien à domicile, dont l’aide pour les travaux ménagers (de 51% à 34% en Ontario). Les personnes âgées, les femmes, les gens vivant seuls ou qui ont besoin d’assistance pour accomplir certaines activités de la vie quotidienne (AVQ) ou activités instrumentales de la vie quotidienne (AIVQ) reçoivent de moins en moins d’aide publique à domicile. En conclusion, le présent système de santé, qui semble se concentrer sur les soins de santé plutôt que sur les services de maintien, ne répond pas aux besoins des personnes âgées. Par ailleurs, le cas ontarien évoque l’impact que peuvent avoir les décisions politiques sur l’utilisation de l’aide à domicile. Les résultats suggèrent qu’entre 1996-1997 et 2005, ce sont surtout les personnes âgées de 18 à 64 ans et ceux nécessitant des soins qui bénéficièrent des programmes publics. Durant la période suivante (2005 à 2009-2010), il semble y avoir un intérêt plus marqué pour les 65 ans et plus dont la proportion d’utilisateurs s’accroît plus rapidement.

Describing Web Service Architectures through Design-by-Contract

Architectural description languages (ADLs) are used to specify a high-level, compositional view of a software application, specifying how a system is to be composed from coarse-grain components. ADLs usually come equipped with a formal dynamic semantics, facilitating specification and analysis of distributed and event-based systems. In this paper, we describe the TrustME, an ADL framework that provides both a process and a structural view of web service-based systems. We use Petri-net descriptions to give a dynamic view of business workflow for web service collaboration. We adapt the approach of Schmidt to define a form of Meyer's design-by-contract for configuring workflow architectures. This serves as a configuration-level means of constructing safer, more robust systems.

An Architectural Description Language for Enterprise Computing

Architectural description languages (ADLs) are used to specify high-level, compositional view of a software application. ADLs usually come equipped with a rigourous state-transition style semantics, facilitating specification and analysis of distributed and event-based systems. However, enterprise system architectures built upon newer middleware (implementations of Java’s EJB specification, or Microsoft’s COM+/ .NET) require additional expressive power from an ADL. The TrustME ADL is designed to meet this need. In this paper, we describe several aspects of TrustME that facilitate specification and anlysis of middleware-based architectures for the enterprise.

Trust-By-Contract: Modelling, Analysing and Predicting Behaviour in Software Architectures

Architecture description languages (ADLs) are used to specify high-level, compositional views of a software application. ADL research focuses on software composed of prefabricated parts, so-called software components. ADLs usually come equipped with rigorous state-transition style semantics, facilitating verification and analysis of specifications. Consequently, ADLs are well suited to configuring distributed and event-based systems. However, additional expressive power is required for the description of enterprise software architectures – in particular, those built upon newer middleware, such as implementations of Java’s EJB specification, or Microsoft’s COM+/.NET. The enterprise requires distributed software solutions that are scalable, business-oriented and mission-critical. We can make progress toward attaining these qualities at various stages of the software development process. In particular, progress at the architectural level can be leveraged through use of an ADL that incorporates trust and dependability analysis. Also, current industry approaches to enterprise development do not address several important architectural design issues. The TrustME ADL is designed to meet these requirements, through combining approaches to software architecture specification with rigorous design-by-contract ideas. In this paper, we focus on several aspects of TrustME that facilitate specification and analysis of middleware-based architectures for trusted enterprise computing systems.

The effect of resistance training on functional capacity and quality of life in individuals with high and low number of metabolic risk factors

OBJECTIVE—There are limited data on the effects of resistance training on the capacity to perform activities of daily living (ADLs) and quality of life (QoL) for individuals with a high number of metabolic risk factors (HiMF). In this study, we examined the effect of resistance training on the capacity to perform ADLs and QoL in individuals with HiMF and compared any benefits with individuals with a low number of metabolic risk factors (LoMF).

RESEARCH DESIGN AND METHODS—Fifty-five untrained individuals, aged 50.8 ± 6.5 years, were randomized to four groups: HiMF training (HiMFT), HiMF control, LoMF training (LoMFT), and LoMF control. At baseline and after 10 weeks of resistance training, participants underwent anthropometric measurements and assessments of aerobic power (Vo2peak), muscle strength, capacity to perform ADLs, and a self-perceived QoL questionnaire. A repeated-measures ANOVA was used to examine the effect of training over time among groups.

RESULTS—Training increased lean body mass in both HiMFT (P = 0.03) and LoMFT (P = 0.03) groups. Total fat content and Vo2peak improved in the LoMFT group only. Muscle strength improved in both training groups (P < 0.01). Time to complete ADLs was reduced by 8.8% in the LoMFT group (P < 0.01) and 9.7% in the HiMF group (P < 0.01). Only the HiMFT group reported improvement in QoL.

CONCLUSIONS— Resistance training improved muscle strength and the capacity to perform ADLs in individuals with HiMF and LoMF. Resistance training improved QoL for the HiMF group, and this result was independent of changes in body fat content or aerobic power. Longer training regimens may be needed to improve QoL in individuals with LoMF.

Human behavior recognition with generic exponential family duration modeling in the hidden semi-Markov model

The ability to learn and recognize human activities of daily living (ADLs) is important in building pervasive and smart environments. In this paper, we tackle this problem using the hidden semi-Markov model. We discuss the state-of-the-art duration modeling choices and then address a large class of exponential family distributions to model state durations. Inference and learning are efficiently addressed by providing a graphical representation for the model in terms of a dynamic Bayesian network (DBN). We investigate both discrete and continuous distributions from the exponential family (Poisson and Inverse Gaussian respectively) for the problem of learning and recognizing ADLs. A full comparison between the exponential family duration models and other existing models including the traditional multinomial and the new Coxian are also presented. Our work thus completes a thorough investigation into the aspect of duration modeling and its application to human activities recognition in a real-world smart home surveillance scenario.

Hierarchical semi-markov conditional random fields for recursive sequential data

Inspired by the hierarchical hidden Markov models (HHMM), we present the hierarchical semi-Markov conditional random field (HSCRF), a generalisation of embedded undirected Markov chains to model complex hierarchical, nested Markov processes. It is parameterised in a discriminative framework and has polynomial time algorithms for learning and inference. Importantly, we develop efficient algorithms for learning and constrained inference in a partially-supervised setting, which is important issue in practice where labels can only be obtained sparsely. We demonstrate the HSCRF in two applications: (i) recognising human activities of daily living (ADLs) from indoor surveillance cameras, and (ii) noun-phrase chunking. We show that the HSCRF is capable of learning rich hierarchical models with reasonable accuracy in both fully and partially observed data cases.

Efficient Coxian duration modelling for activity recognition in smart environment with the hidden semi-Markov model

In this paper, we exploit the discrete Coxian distribution and propose a novel form of stochastic model, termed as the Coxian hidden semi-Makov model (Cox-HSMM), and apply it to the task of recognising activities of daily living (ADLs) in a smart house environment. The use of the Coxian has several advantages over traditional parameterization (e.g. multinomial or continuous distributions) including the low number of free parameters needed, its computational efficiency, and the existing of closed-form solution. To further enrich the model in real-world applications, we also address the problem of handling missing observation for the proposed Cox-HSMM. In the domain of ADLs, we emphasize the importance of the duration information and model it via the Cox-HSMM. Our experimental results have shown the superiority of the Cox-HSMM in all cases when compared with the standard HMM. Our results have further shown that outstanding recognition accuracy can be achieved with relatively low number of phases required in the Coxian, thus making the Cox-HSMM particularly suitable in recognizing ADLs whose movement trajectories are typically very long in nature.

The development of the Melbourne low vision ADL index : a measure of vision disability

PURPOSE. To develop a new test of activities of daily living (ADLs) appropriate for the low-vision population: the Melbourne Low-Vision ADL Index (MLVAI).

METHODS. The MLVAI was designed as a desk-based clinical assessment, comprising 18 observed items on complex ADLs in part (a) and 9 questions on broad self-care ADLs in part (b). Each item was rated on a five-level descriptive scale from 0 to 4, based on independence, speed, and accuracy of performance. It was designed to be administered under standardized conditions with regard to the instructions, illumination, and working distances. The validity and reliability of the new MLVAI was determined for 122 subjects who were representative of the general low-vision population, in a cross-sectional study.

RESULTS. Two items were found to be redundant and were eliminated from the test. Thus, the final test comprised 25 items, with 100 being the highest possible score. Cronbach’s α indicated an internal reliability of 0.96, and an intraclass correlation coefficient indicated an overall reliability of 0.95. The SE of measurement was 4.5. According to Spearman’s correlation coefficient, the test–retest reliability was 0.94 (P < 0.001), and the interpractitioner reliability for five different pairs of practitioners was 0.90 or higher (P < 0.001). With regard to validity, there was a moderately high correlation with vision impairment (r = −0.68, P < 0.001). Using Rasch analysis, content validity was also demonstrated by good separation indexes (4.70 and 9.88) and high reliability scores (0.96 and 0.99) for the person and items parameters, respectively. Separate calculation of indexes and reliability scores for parts (a) and (b) indicated high content validity and reliability of each part. However, the separation indexes and reliability scores were higher for part (a) than for part (b). The correlation coefficient for part (a) and part (b) was 0.68.

CONCLUSIONS. The MLVAI is a highly valid and reliable standardized test of ADL performance for the general low-vision population. It may be used to assess patients with low vision and has the potential to be used as a measure of low-vision rehabilitation outcomes.