Estimating the remaining useful life (RUL) of an asset lies at the heart of prognostics and health management (PHM) of many operations-critical industries such as aviation. Mod- ern methods of RUL estimation adopt techniques from deep learning (DL). However, most of these contemporary tech- niques deliver only single-point estimates for the RUL without reporting on the confidence of the prediction. This practice usually provides overly confident predictions that can have severe consequences in operational disruptions or even safety. To address this issue, we propose a technique for uncertainty quantification (UQ) based on Bayesian deep learning (BDL). The hyperparameters of the framework are tuned using a novel bi-objective Bayesian optimization method with objectives the predictive performance and predictive uncertainty. The method also integrates the data pre-processing steps into the hyperparameter optimization (HPO) stage, models the RUL as a Weibull distribution, and returns the survival curves of the monitored assets to allow informed decision-making. We vali- date this method on the widely used C-MAPSS dataset against a single-objective HPO baseline that aggregates the two ob- jectives through the harmonic mean (HM). We demonstrate the existence of trade-offs between the predictive performance and the predictive uncertainty and observe that the bi-objective HPO returns a larger number of hyperparameter configurations compared to the single-objective baseline. Furthermore, we see that with the proposed approach, it is possible to configure models for RUL estimation that exhibit better or comparable performance to the single-objective baseline when validated on the test sets.
DOCUMENT
Remaining Useful Life (RUL) estimation is directly related with the application of predictive maintenance. When RUL estimation is performed via data-driven methods and Artificial Intelligence algorithms, explainability and interpretability of the model are necessary for trusted predictions. This is especially important when predictive maintenance is applied to gas turbines or aeroengines, as they have high operational and maintenance costs, while their safety standards are strict and highly regulated. The objective of this work is to study the explainability of a Deep Neural Network (DNN) RUL prediction model. An open-source database is used, which is composed by computed measurements through a thermodynamic model for a given turbofan engine, considering non-linear degradation and data points for every second of a full flight cycle. First, the necessary data pre-processing is performed, and a DNN is used for the regression model. The selection of its hyper-parameters is done using random search and Bayesian optimisation. Tests considering the feature selection and the requirements of additional virtual sensors are discussed. The generalisability of the model is performed, showing that the type of faults as well as the dominant degradation has an important effect on the overall accuracy of the model. The explainability and interpretability aspects are studied, following the Local Interpretable Model-agnostic Explanations (LIME) method. The outcomes are showing that for simple data sets, the model can better understand physics, and LIME can give a good explanation. However, as the complexity of the data increases, both the accuracy of the model drops but also LIME seems to have difficulties in giving satisfactory explanations.
DOCUMENT
Background: Hypothyroidism is a common endocrine disorder and the standard treatment is replacement therapy with levothyroxine (LT4). Although many hypothyroid patients improve upon treatment with LT4, a proportion seems to experience residual hypothyroid complaints despite treatment, even when plasma TSH and FT4 are within reference ranges. Methods: Using an on-line survey we investigated 1. the health-related quality of life (QoL) (ThyPRO), 2. the activities of daily living (SF-36), 3. hypothyroid-related symptoms (ThySHI) in diagnosed, treated hypothyroid patients (>18 years, treated >6 months) and control persons (without thyroid disease, >18 years). In patients, the time course of symptoms from diagnosis until 3 years was asked (retrospectively, ThySHI). Patients and control persons were recruited by e-mails from patient organizations, posters in pharmacies and health centers and Twitter/Facebook. For data analysis (ThyPRO, 0-100 scale, t-test; daily functioning, 1-5 scale and ThySHI 0-3 scale, Mann-Whitney; time course symptoms, Friedmann-Dunnett; confounding factors, ANCOVA) IBM SPSS 24 was used. Results: In this cohort consisted of 1667 patients (mean duration of illness 12.2 ± SD 9.9 years) and 275 controls. Treated hypothyroid patients had 1. a significant decrease in health-related QoL and all domains (fatigue, vitality, cognition, anxiety, depressivity, emotional susceptibility, social life, daily life), as compared to controls (mean total QoL 39.9 vs 19.1 resp. and all domains p<0.001), 2. Significantly more impairment with activities of daily living (p<0.001), and 3. significantly higher scores for symptoms related to hypothyroidism, as compared to control persons (all p<0.01). Symptoms generally decreased after 3 years of treatment, with fatigue, reduce daily functioning, coldness, muscle pain/cramps and being overweight as the most intense residual complaints. Many patients (78.5%) reported having complaints despite taking thyroid medication and reported not feeling well (77.8%) while their blood values were within range. TSH level, age, gender and duration of illness did not significantly affect total QoL, whereas the M3 comorbidity index did. Desiccated thyroid hormone users (9.4%) had a significantly better mean total QoL than LT4 users (90.5%) (36,0 vs 40.6, p=0.003). Conclusions: Persistent complaints, such as reduced health-related quality of life, reduced daily functioning, and residual hypothyroid related symptoms, are common in this group of hypothyroid patients despite replacement therapy. Caregivers should be aware that persistent complaints can be present in treated hypothyroid patients, despite following current guidelines, and that these remaining symptoms may affect their quality of life and daily functioning.
DOCUMENT
Abstract Purpose To determine the predictive value of quality of life for mortality at the domain and item levels. Methods This longitudinal study was carried out in a sample of 479 Dutch people aged 75 years or older living independently, using a follow-up of 7 years. Participants completed a self-report questionnaire. Quality of life was assessed with the WHOQOL-BREF, including four domains: physical health, psychological, social relationships, and environment. The municipality of Roosendaal (a town in the Netherlands) indicated the dates of death of the individuals. Results Based on mean, all quality of life domains predicted mortality adjusted for gender, age, marital status, education, and income. The hazard ratios ranged from 0.811 (psychological) to 0.933 (social relationships). The areas under the curve (AUCs) of the four domains were 0.730 (physical health), 0.723 (psychological), 0.693 (social relationships), and 0.700 (environment). In all quality of life domains, at least one item predicted mortality (adjusted). Conclusion Our study showed that all four quality of life domains belonging to the WHOQOL-BREF predict mortality in a sample of Dutch community-dwelling older people using a follow-up period of 7 years. Two AUCs were above threshold (psychological, physical health). The findings offer health care and welfare professionals evidence for conducting interventions to reduce the risk of premature death.
DOCUMENT
As the population ages, the risk of becoming malnourished increases. Research has shown that poor oral health can be a risk factor for malnutrition in institutionalized elderly. However, it remains unclear whether oral health problems, edentulousness and health-related quality of life also pose a risk for malnutrition in community-dwelling older adults. In this cross-sectional observational study, 1325 community-living elderly (≥75 years) were asked to complete questionnaires regarding nutritional status, oral status (edentulous, remaining teeth, or implant-supported overdentures), oral health problems, health-related quality of life (HRQoL), frailty, activities of daily living (ADL) and complexity of care needs. Univariate and multivariate logistic regression analyses were performed with nutritional status as dependent variable. Of the respondents, 51% (n = 521) were edentulous, 38.8% (n = 397) had remaining teeth and 10.2% (n = 104) had an implant-supported overdenture. Elderly with complex care needs were malnourished most frequently, followed by frail and robust elderly (10%, 4.5% and 2.9%, respectively). Malnourished elderly reported more frequent problems with chewing and speech when compared with well-nourished elderly (univariate analysis). However, multivariate analysis did not show an association between malnutrition and oral health problems and edentulousness, although HRQoL was associated with malnutrition (odds ratio (OR) 0.972, confidence interval (CI) 0.951–0.955). Based on the results of this cross-sectional study, it can be concluded that poor HRQoL is significantly associated with malnutrition; however, edentulousness and oral health problems are not.
DOCUMENT
Intra-ocular straylight can cause decreased visual functioning, and it may cause diminished vision-related quality of life (VRQOL). This cross-sectional population-based study investigates the association between straylight and VRQOL in middle-aged and elderly individuals. Multivariable linear regression analyses were used to assess the association between straylight modeled continuously and cutoff at the recommended fitness-to-drive value, straylight ≥ 1.4 log(s), and VRQOL. The study showed that participants with normal straylight values, straylight ≤ 1.4 log(s), rated their VRQOL slightly better than those with high straylight values (straylight ≥ 1.4 log(s)). Furthermore, multivariable regression analysis revealed a borderline statistical significant association (p = .06) between intra-ocular straylight and self-reported VRQOL in middle-aged and elderly individuals. The association between straylight and self-reported VRQOL was not influenced by the status of the intra-ocular lens (natural vs. artificial intra-ocular lens after cataract extraction) or the number of (instrumental) activities of daily living that were reported as difficult for the elderly individuals.
DOCUMENT
This papers presents some ideas to use so-called software agents as a software representation of a product not only during manufacturing but also during the whole life cycle of the product. Software agents are autonomous entities capable of collecting useful information about products. By their design and capabilities software agents fit well in the concept of ubiquitous computing. We use these agents in our newly developed manufacturing process. This paper discusses further use of agent technology.
DOCUMENT
As migrant populations age, the care system is confronted with the question how to respond to care needs of an increasingly diverse population of older adults. We used qualitative intersectional analysis to examine differential preferences and experiences with care at the end of life of twenty-five patients and their relatives from Suriname, Morocco and Turkey living in The Netherlands. Our analysis focused on the question how–in light of impairment–ethnicity, religion and gender intersect to create differences in social position that shape preferences and experiences related to three main themes: place of care at the end of life; discussing prognosis, advance care, and end-of-life care; and, end-of-life decision-making. Our findings show that belonging to an ethnic or religious minority brings forth concerns about responsive care. In the nursing home, patients’ minority position and the interplay thereof with gender make it difficult for female patients to request and receive responsive care. Patients with a strong religious affiliation prefer to discuss diagnosis but not prognosis. These preferences are at interplay with factors related to socioeconomic status. The oversight of this variance hampers responsive care for patients and relatives. Preferences for discussion of medical aspects of care are subject to functional impairment and faith. Personal values and goals often remain unexpressed. Lastly, preferences regarding medical end-of-life decisions are foremost subject to religious affiliation and associated moral values. Respondents’ impairment and limited Dutch language proficiency requires their children to be involved in decision-making. Intersecting gendered care roles determine that mostly daughters are involved. Considering the interplay of aspects of social identity and their effect on social positioning, and pro-active enquiry into values, goals and preferences for end-of-life care of patients and their relatives are paramount to achieve person centred and family-oriented care responsive to the needs of diverse communities.
DOCUMENT
Discussions on policy and management initiatives to facilitate individuals throughout working careers take place without sufficient insight into how career paths are changing, how these changes are related to a modernization of life course biographies, and whether this leads to increased labour market transitions. This paper asks how new, flexible labour market patterns can best be analyzed using an empirical, quantitative approach. The data used are from the career module of the Panel Study of Belgian Households (PSBH). This module, completed by almost 4500 respondents consists of retrospective questions tracing lengthy and even entire working life histories. To establish any changes in career patterns over such extended periods of time, we compare two evolving methodologies: Optimal Matching Analysis (OMA) and Latent Class Regression Analysis (LCA). The analyses demonstrate that both methods show promising potential in discerning working life typologies and analyzing sequence trajectories. However, particularities of the methods demonstrate that not all research questions are suitable for each method. The OMA methodology is appropriate when the analysis concentrates on the labour market statuses and is well equipped to make clear and interpretable differentiations if there is relative stability in career paths during the period of observation but not if careers become less stable. Latent Class has the strength of adopting covariates in the clustering allowing for more historically connected types than the other methodology. The clustering is denser and the technique allows for more detailed model fitting controls than OMA. However, when incorporating covariates in a typology, the possibilities of using the typology in later, causal, analyses is somewhat reduced.
MULTIFILE
Research has shown that some 30% of total care needs in people with late-life depression (LLD) are unmet. It is not known to what extent patients actually don’t receive any care for these needs or consider the care to be insufficient and their satisfaction with the provided care. Results: In 67% of patients, at least one unmet need was ascertained. In most cases (80%) care was actually provided for those needs by professionals and/or informal caregivers. Patients were satisfied with the care delivered for 81% of the reported care needs. Satisfaction was lowest for social care needs (67%). For six specific care needs it was demonstrated that dissatisfied patients were significantly more depressed than satisfied patients. Conclusion: Even though patients might receive care for certain needs, this does not mean that their needs are met. A substantial proportion of patients with LDD feel that they need additional help for unmet needs.
DOCUMENT
