Matthew Carr

A model for residual life prediction based on Brownian motion with an adaptive drift

Abstract A degradation model is presented in this paper for the prediction of the residual life using an adapted Brownian motion-based approach with a drifting parameter. This model differs from other Brownian motion-based approaches in that the drifting parameter of the degradation process is adapted to the history of monitored information. This adaptation is performed by Kalman filtering. We also use a threshold distribution instead of the usual single threshold line which is sometime difficult to obtain in practice. We demonstrate the model using some examples and show that the model performs reasonably well and has a better prediction ability than the standard Brownian motion-based model. The model is then fitted to the data generated from a simulator using the expectation–maximization algorithm. We also fit a standard Brownian motion-based model to the same data to compare the difference and performance. The result shows that the adapted model performs better in terms of certain test statistics and the total mean square errors.

Baseline Bleeding Risk and Arterial Access Site Practice in Relation to Procedural Outcomes After Percutaneous Coronary Intervention

BACKGROUND Transradial access (TRA) has been associated with reduced access site-related bleeding complications and mortality after percutaneous coronary intervention (PCI). It is unclear, however, whether these observed benefits are influenced by baseline bleeding risk. OBJECTIVES This study investigated the relationship between baseline bleeding risk, TRA utilization, and procedure-related outcomes in patients undergoing PCI enrolled in the British Cardiovascular Intervention Society database. METHODS Baseline bleeding risk was calculated by using modified Mehran bleeding risk scores in 348,689 PCI procedures performed between 2006 and 2011. Four categories for bleeding risk were defined for the modified Mehran risk score (MMRS): low (<10), moderate (10 to 14), high (15 to 19), and very high (≥20). The impact of baseline bleeding risk on 30-day mortality and its relationship with access site were studied. RESULTS TRA was independently associated with a 35% reduction in 30-day mortality risk (odds ratio [OR]: 0.65 [95% confidence interval (CI): 0.59 to 0.72]; p < 0.0001), with the magnitude of mortality reduction related to baseline bleeding risk (MMRS <10, OR: 0.73 [95% CI: 0.62 to 0.86]; MMRS ≥20, OR: 0.53 [95% CI: 0.47 to 0.61]). In patients with an MMRS <10, TRA was used in 71,771 (43.2%) of 166,083 PCI procedures; TRA was used in 8,655 (40.1%) of 21,559 PCI procedures in patients with an MMRS ≥20, illustrating that TRA was used less in those at highest risk from bleeding complications (p < 0.0001). CONCLUSIONS TRA was independently associated with reduced 30-day mortality, and the magnitude of this effect was related to baseline bleeding risk; those at highest risk of bleeding complications gained the greatest benefit from adoption of TRA during PCI.

An approximate algorithm for prognostic modelling using condition monitoring information

Established condition based maintenance modelling techniques can be computationally expensive. In this paper we propose an approximate methodology using extended Kalman-filtering and condition monitoring information to recursively establish a conditional probability density function for the residual life of a component. The conditional density is then used in the construction of a maintenance/replacement decision model. The advantages of the methodology, when compared with alternative approaches, are the direct use of the often multi-dimensional condition monitoring data and the on-line automation opportunity provided by the computational efficiency of the model that potentially enables the simultaneous condition monitoring and associated inference for a large number of components and monitored variables. The methodology is applied to a vibration monitoring scenario and compared with alternative models using the case data.

Modeling Failure Modes for Residual Life Prediction Using Stochastic Filtering Theory

This paper reports on a theoretical Bayesian modeling development for residual life prediction in the context of condition-based maintenance. At each monitoring point during a components lifetime, the stochastic filter is used to establish a posterior conditional probability density function (PDF) for the residual life. The PDF can then be used in the evaluation of maintenance and replacement decisions. The research documented in this paper extends the modeling principles in accordance with a practical consideration recognized in a number of previous case applications. Many monitoring scenarios provide evidence that the operational components involved may potentially be subject to a number of individual distinct failure modes, rather than a single dominant failure mode as modeled previously. The modeling procedure proposed to handle this scenario is based on the assumption that an individual monitored component will fail according to one of a number of predefined failure modes. Individual stochastic filters are constructed to facilitate the residual life prediction under the influence of each potential failure mode, and the output from each filter is weighted according to the probability that the particular failure mode is the true underlying (unknown) failure mode. The probabilities associated with each failure mode are recursively derived using a Bayesian model, and the condition monitoring information obtained to date. The modeling process is applied to a set of simulated condition monitoring histories incorporating two potential failure modes, and the results are compared with those obtained using a general model with no failure mode assumptions. The results indicate that, when individual failure modes are identifiable using historical data, the modeling process described in the paper could greatly improve residual life prediction capabilities, and prevent the occurrence of costly component failures.

Incidence, clinical management, and mortality risk following self harm among children and adolescents: cohort study in primary care.

Objectives To examine temporal trends in sex and age specific incidence of self harm in children and adolescents, clinical management patterns, and risk of cause specific mortality following an index self harm episode at a young age. Design Population based cohort study. Setting UK Clinical Practice Research Datalink—electronic health records from 674 general practices, with practice level deprivation measured ecologically using the index of multiple deprivation. Patients from eligible English practices were linked to hospital episode statistics (HES) and Office for National Statistics (ONS) mortality records. Participants For the descriptive analytical phases we examined data pertaining to 16 912 patients aged 10-19 who harmed themselves during 2001-14. For analysis of cause specific mortality following self harm, 8638 patients eligible for HES and ONS linkage were matched by age, sex, and general practice with up to 20 unaffected children and adolescents (n=170 274). Main outcome measures In the first phase, temporal trends in sex and age specific annual incidence were examined. In the second phase, clinical management was assessed according to the likelihood of referral to mental health services and psychotropic drug prescribing. In the third phase, relative risks of all cause mortality, unnatural death (including suicide and accidental death), and fatal acute alcohol or drug poisoning were estimated as hazard ratios derived from stratified Cox proportional hazards models for the self harm cohort versus the matched unaffected comparison cohort. Results The annual incidence of self harm was observed to increase in girls (37.4 per 10 000) compared with boys (12.3 per 10 000), and a sharp 68% increase occurred among girls aged 13-16, from 45.9 per 10 000 in 2011 to 77.0 per 10 000 in 2014. Referrals within 12 months of the index self harm episode were 23% less likely for young patients registered at the most socially deprived practices, even though incidences were considerably higher in these localities. Children and adolescents who harmed themselves were approximately nine times more likely to die unnaturally during follow-up, with especially noticeable increases in risks of suicide (deprivation adjusted hazard ratio 17.5, 95% confidence interval 7.6 to 40.5) and fatal acute alcohol or drug poisoning (34.3, 10.2 to 115.7). Conclusions Gaining a better understanding of the mechanisms responsible for the recent apparent increase in the incidence of self harm among early-mid teenage girls, and coordinated initiatives to tackle health inequalities in the provision of services to distressed children and adolescents, represent urgent priorities for multiple public agencies.

A stochastic filtering based data driven approach for residual life prediction and condition based maintenance decision making support

As an efficient means of detecting potential plant failure, condition monitoring is growing popular in industry with millions spent on condition monitoring hardware and software. The use of condition monitoring techniques will generally increase plant availability and reduce downtime costs, but in some cases it will also tend to over-maintaining the plant in question. There is obviously a need for appropriate decision support in plant maintenance planning utilising available condition monitoring information, but compared to the extensive literature on diagnosis, relatively little research has been done on the prognosis side of condition based maintenance. In plant prognosis, a key, but often uncertain quantity to be modelled is the residual life prediction based on available condition information to date. This paper shall focus upon such a residual life prediction of the monitored items in condition based maintenance and review the recent developments in modelling residual life prediction using stochastic filtering. We first demonstrate the role of residual life prediction in condition based maintenance decision making, which highlights the need for such a prediction. We then discuss in detail the basic filtering model we used for residual life prediction and the extensions we made. We finally present briefly the result of the comparative studies between the filtering based model and other models using empirical data. The results show that the filtering based approach is the best in terms of prediction accuracy and cost effectiveness.

Incorporating the potential for human error in maintenance models

The mathematics of delay-time modelling of inspection maintenance is extended to incorporate the existence of human error in the form of fault injection during the inspection process. After briefly discussing the basic delay-time model, modifications are introduced to model maintenance scenarios incorporating human error injected defects within the inspection maintenance process. The effects of human error are investigated with the emphasis on its representation and on the assessment of consequences, the objective being to provide a means of determining the cost of human error and to thereby aid corrective decision-making.

The predictive ability of blood pressure in elderly trial patients

Objectives: To assess the impact of the blood pressure (BP) profile on cardiovascular risk in the Medical Research Council (UK) elderly trial; investigate whether the effects of hypertensive drugs in reducing event rates are solely a product of systolic pressure reduction. Methods: Using longitudinal BP data from 4396 hypertensive patients, the general trend over time was estimated using a first-stage multilevel model. We then investigated how BP acted alongside other BP-related covariates in a second-stage ‘time-to-event’ statistical model, assessing risk for stroke events and coronary heart disease (CHD). Differences in outcome prediction between diuretic, &bgr;-blocker and placebo treatment arms were investigated. Results: The &bgr;-blocker arm experienced comparatively poor control of current SBP, episodic peaks and variability in BP levels. After adjusting for the mean level, variability in SBP over time was significant: risk ratio was 1.15 [95% confidence interval (CI): 1.01–1.31] across all patients for stroke events. The risk ratio for current SBP was 1.36 (95% CI: 1.16–1.58). Current DBP and variability in DBP also predicted stroke independently: risk ratios was 1.43 and 1.18, respectively. The risk factors exhibited weaker associations with CHD risk; only the highest measured value and variability in SBP showed a statistically significant association: risk ratios were 1.26 and 1.16, respectively. Conclusion: Individual risk characterization could be augmented with additional prognostic information, besides current SBP, including current diastolic pressure, temporal variability over and above general trends and historical measurements.

Examining Trends in Type 2 Diabetes Incidence, Prevalence and Mortality in the UK between 2004 and 2014

Contemporary data describing type 2 diabetes prevalence, incidence and mortality are limited. We aimed to (1) estimate annual incidence and prevalence rates of type 2 diabetes in the UK between 2004 and 2014, (2) examine relationships between observed rates with age, gender, socio‐economic status and geographic region, and (3) assess how temporal changes in incidence and all‐cause mortality rates influence changes in prevalence.

Premature Mortality Among Patients Recently Discharged From Their First Inpatient Psychiatric Treatment

Importance Nationwide cohorts provide sufficient statistical power for examining premature, cause-specific mortality in patients recently discharged from inpatient psychiatric services. Objective To investigate premature mortality in a nationwide cohort of patients recently discharged from inpatient psychiatric treatment at ages 15 to 44 years. Design, Setting, and Participants This single-cohort design included all persons born in Denmark (N = 1 683 385) between January 1, 1967, and December 31, 1996. Exactly 48 599 of these Danish residents were discharged from an inpatient psychiatric unit or ward on or after their 15th birthday, which took place during this study’s observation period from January 1, 1982, through December 31, 2011. This group of patients was followed up beginning on their 15th birthday until their death, emigration, or December 31, 2011, whichever came first. Individuals discharged from inpatient psychiatric care at least once before their 15th birthday (n = 5882) were excluded from the study. All data were obtained from the Danish Civil Registration System, Psychiatric Central Research Register, and Register of Causes of Death. Data analysis took place between February 1, 2016, and December 10, 2016. Main Outcomes and Measures Incidence rates and incidence rate ratios (IRRs) for all-cause mortality and for an array of unnatural and natural causes of death among patients recently discharged from an inpatient psychiatric unit vs persons not admitted to a psychiatric facility. Primary analysis considered risk within the year of first discharge. Results Of the 48 599 discharged patients who were included in the study, 25 006 (51.4%) were female, 35 660 (73.4%) were aged 15 to 29 years, and 33 995 (70.0%) had a length of stay of 30 days or less. Compared with persons not admitted, patients discharged had an elevated risk for all-cause mortality within 1 year (IRR, 16.2; 95% CI, 14.5-18.0). The relative risk for unnatural death (IRR, 25.0; 95% CI, 22.0-28.4) was much higher than for natural death (IRR, 8.6; 95% CI, 7.0-10.7). The highest IRR found was for suicide at 66.9 (95% CI, 56.4-79.4), followed by alcohol-related death at 42.0 (95% CI, 26.6-66.1). Among the psychiatric diagnostic categories assessed, psychoactive substance abuse conferred the highest risk for all-cause mortality (IRR, 24.8; 95% CI, 21.0-29.4). Across the array of cause-specific outcomes examined, risk of premature death during the first year after discharge was markedly higher than the risk of death beyond the first year of discharge. Conclusions and Relevance Clinicians may help protect patients after discharge by serving as a liaison between primary and secondary health services to ensure they are receiving holistic care. Early intervention programs for drug and alcohol misuse could substantially decrease the greatly elevated mortality risk among these patients.