Which screening strategy using BMD measurements would be most cost effective for hip fracture prevention in elderly women? A decision analysis based on a Markov model.

INTRODUCTION: Hip fractures are responsible for excessive mortality, decreasing the 5-year survival rate by about 20%. From an economic perspective, they represent a major source of expense, with direct costs in hospitalization, rehabilitation, and institutionalization. The incidence rate sharply increases after the age of 70, but it can be reduced in women aged 70-80 years by therapeutic interventions. Recent analyses suggest that the most efficient strategy is to implement such interventions in women at the age of 70 years. As several guidelines recommend bone mineral density (BMD) screening of postmenopausal women with clinical risk factors, our objective was to assess the cost-effectiveness of two screening strategies applied to elderly women aged 70 years and older. METHODS: A cost-effectiveness analysis was performed using decision-tree analysis and a Markov model. Two alternative strategies, one measuring BMD of all women, and one measuring BMD only of those having at least one risk factor, were compared with the reference strategy "no screening". Cost-effectiveness ratios were measured as cost per year gained without hip fracture. Most probabilities were based on data observed in EPIDOS, SEMOF and OFELY cohorts. RESULTS: In this model, which is mostly based on observed data, the strategy "screen all" was more cost effective than "screen women at risk." For one woman screened at the age of 70 and followed for 10 years, the incremental (additional) cost-effectiveness ratio of these two strategies compared with the reference was 4,235 euros and 8,290 euros, respectively. CONCLUSION: The results of this model, under the assumptions described in the paper, suggest that in women aged 70-80 years, screening all women with dual-energy X-ray absorptiometry (DXA) would be more effective than no screening or screening only women with at least one risk factor. Cost-effectiveness studies based on decision-analysis trees maybe useful tools for helping decision makers, and further models based on different assumptions should be performed to improve the level of evidence on cost-effectiveness ratios of the usual screening strategies for osteoporosis.

Decision analysis for the genotype designation in low-template-DNA profiles

What genotype should the scientist specify for conducting a database search to try to find the source of a low-template-DNA (lt-DNA) trace? When the scientist answers this question, he or she makes a decision. Here, we approach this decision problem from a normative point of view by defining a decision-theoretic framework for answering this question for one locus. This framework combines the probability distribution describing the uncertainty over the trace's donor's possible genotypes with a loss function describing the scientist's preferences concerning false exclusions and false inclusions that may result from the database search. According to this approach, the scientist should choose the genotype designation that minimizes the expected loss. To illustrate the results produced by this approach, we apply it to two hypothetical cases: (1) the case of observing one peak for allele xi on a single electropherogram, and (2) the case of observing one peak for allele xi on one replicate, and a pair of peaks for alleles xi and xj, i ≠ j, on a second replicate. Given that the probabilities of allele drop-out are defined as functions of the observed peak heights, the threshold values marking the turning points when the scientist should switch from one designation to another are derived in terms of the observed peak heights. For each case, sensitivity analyses show the impact of the model's parameters on these threshold values. The results support the conclusion that the procedure should not focus on a single threshold value for making this decision for all alleles, all loci and in all laboratories.

The effectiveness of brief alcohol interventions in primary care settings: a systematic review.

Issues. Numerous studies have reported that brief interventions delivered in primary care are effective in reducing excessive drinking. However, much of this work has been criticised for being clinically unrepresentative. This review aimed to assess the effectiveness of brief interventions in primary care and determine if outcomes differ between efficacy and effectiveness trials. Approach. A pre-specified search strategy was used to search all relevant electronic databases up to 2006. We also hand-searched the reference lists of key articles and reviews. We included randomised controlled trials (RCT) involving patients in primary care who were not seeking alcohol treatment and who received brief intervention. Two authors independently abstracted data and assessed trial quality. Random effects meta-analyses, subgroup and sensitivity analyses and meta-regression were conducted. Key Findings. The primary meta-analysis included 22 RCT and evaluated outcomes in over 5800 patients. At 1 year follow up, patients receiving brief intervention had a significant reduction in alcohol consumption compared with controls [mean difference: -38 g week(-1), 95%CI (confidence interval): -54 to -23], although there was substantial heterogeneity between trials (I(2) = 57%). Subgroup analysis confirmed the benefit of brief intervention in men but not in women. Extended intervention was associated with a non-significantly increased reduction in alcohol consumption compared with brief intervention. There was no significant difference in effect sizes for efficacy and effectiveness trials. Conclusions. Brief interventions can reduce alcohol consumption in men, with benefit at a year after intervention, but they are unproven in women for whom there is insufficient research data. Longer counselling has little additional effect over brief intervention. The lack of differences in outcomes between efficacy and effectiveness trials suggests that the current literature is relevant to routine primary care.

The use of classification and regression trees to predict the likelihood of seasonal influenza.

Background Individual signs and symptoms are of limited value for the diagnosis of influenza. Objective To develop a decision tree for the diagnosis of influenza based on a classification and regression tree (CART) analysis. Methods Data from two previous similar cohort studies were assembled into a single dataset. The data were randomly divided into a development set (70%) and a validation set (30%). We used CART analysis to develop three models that maximize the number of patients who do not require diagnostic testing prior to treatment decisions. The validation set was used to evaluate overfitting of the model to the training set. Results Model 1 has seven terminal nodes based on temperature, the onset of symptoms and the presence of chills, cough and myalgia. Model 2 was a simpler tree with only two splits based on temperature and the presence of chills. Model 3 was developed with temperature as a dichotomous variable (≥38°C) and had only two splits based on the presence of fever and myalgia. The area under the receiver operating characteristic curves (AUROCC) for the development and validation sets, respectively, were 0.82 and 0.80 for Model 1, 0.75 and 0.76 for Model 2 and 0.76 and 0.77 for Model 3. Model 2 classified 67% of patients in the validation group into a high- or low-risk group compared with only 38% for Model 1 and 54% for Model 3. Conclusions A simple decision tree (Model 2) classified two-thirds of patients as low or high risk and had an AUROCC of 0.76. After further validation in an independent population, this CART model could support clinical decision making regarding influenza, with low-risk patients requiring no further evaluation for influenza and high-risk patients being candidates for empiric symptomatic or drug therapy.

What do we gain from simplicity versus complexity in species distribution models?

Species distribution models (SDMs) are widely used to explain and predict species ranges and environmental niches. They are most commonly constructed by inferring species' occurrence-environment relationships using statistical and machine-learning methods. The variety of methods that can be used to construct SDMs (e.g. generalized linear/additive models, tree-based models, maximum entropy, etc.), and the variety of ways that such models can be implemented, permits substantial flexibility in SDM complexity. Building models with an appropriate amount of complexity for the study objectives is critical for robust inference. We characterize complexity as the shape of the inferred occurrence-environment relationships and the number of parameters used to describe them, and search for insights into whether additional complexity is informative or superfluous. By building 'under fit' models, having insufficient flexibility to describe observed occurrence-environment relationships, we risk misunderstanding the factors shaping species distributions. By building 'over fit' models, with excessive flexibility, we risk inadvertently ascribing pattern to noise or building opaque models. However, model selection can be challenging, especially when comparing models constructed under different modeling approaches. Here we argue for a more pragmatic approach: researchers should constrain the complexity of their models based on study objective, attributes of the data, and an understanding of how these interact with the underlying biological processes. We discuss guidelines for balancing under fitting with over fitting and consequently how complexity affects decisions made during model building. Although some generalities are possible, our discussion reflects differences in opinions that favor simpler versus more complex models. We conclude that combining insights from both simple and complex SDM building approaches best advances our knowledge of current and future species ranges.

Neuroimaging techniques in Alzheimer's disease

Neuroimaging techniques provide valuable tools for diagnosing Alzheimer's disease (AD), monitoring disease progression and evaluating responses to treatment. There is currently a wide array of techniques available including computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), and, for recording electrical brain activity, electroencephalography (EEG). The choice of technique depends on the contrast between tissues of interest, spatial resolution, temporal resolution, requirements for functional data and the probable number of scans required. For example, while PET, CT and MRI can be used to differentiate between AD and other dementias, MRI is safer and provides better contrast of soft tissues. Neuroimaging is a technique spanning many disciplines and requires effective communication between doctors requesting a scan of a patient or group of patients and those with technical expertise. Consideration and discussion of the most suitable type of scan and the necessary settings to achieve the best results will help ensure appropriate techniques are chosen and used effectively. Neuroimaging techniques are currently expanding understanding of the structural and functional changes that occur in dementia. Further research may allow identification of early neurological signs ofAD, before clinical symptoms are evident, providing the opportunity to test preventative therapies. CombiningMRI and machine learning techniques may be a powerful approach to improve diagnosis ofAD and to predict clinical outcomes.

Factors associated with favorable drinking outcome 12 months after hospitalization in a prospective cohort study of inpatients with unhealthy alcohol use.

BACKGROUND: Prevalence of unhealthy alcohol use among medical inpatients is high. OBJECTIVE: To characterize the course and outcomes of unhealthy alcohol use, and factors associated with these outcomes. DESIGN: Prospective cohort study. PARTICIPANTS: A total of 287 medical inpatients with unhealthy alcohol use. MAIN MEASURES: At baseline and 12 months later, consumption and alcohol-related consequences were assessed. The outcome of interest was a favorable drinking outcome at 12 months (abstinence or drinking "moderate" amounts without consequences). The independent variables evaluated included demographics, physical/sexual abuse, drug use, depressive symptoms, alcohol dependence, commitment to change (Taking Action), spending time with heavy-drinking friends and receipt of alcohol treatment (after hospitalization). Adjusted regression models were used to evaluate factors associated with a favorable outcome. KEY RESULTS: Thirty-three percent had a favorable drinking outcome 1 year later. Not spending time with heavy-drinking friends [adjusted odds ratio (AOR) 2.14, 95% CI: 1.14-4.00] and receipt of alcohol treatment [AOR (95% CI): 2.16(1.20-3.87)] were associated with a favorable outcome. Compared to the first quartile (lowest level) of Taking Action, subjects in the second, third and highest quartiles had higher odds of a favorable outcome [AOR (95% CI): 3.65 (1.47, 9.02), 3.39 (1.38, 8.31) and 6.76 (2.74, 16.67)]. CONCLUSIONS: Although most medical inpatients with unhealthy alcohol use continue drinking at-risk amounts and/or have alcohol-related consequences, one third are abstinent or drink "moderate" amounts without consequences 1 year later. Not spending time with heavy-drinking friends, receipt of alcohol treatment and commitment to change are associated with this favorable outcome. This can inform efforts to address unhealthy alcohol use among patients who often do not seek specialty treatment.

The effectiveness of brief alcohol interventions in primary care settings: A systematic review

Numerous studies have reported that brief interventions delivered in primary care are effective in reducing excessive drinking. However, much of this work has been criticised for being clinically unrepresentative. This review aimed to assess the effectiveness of brief interventions in primary care and determine if outcomes differ between efficacy and effectiveness trials. Approach. A pre-specified search strategy was used to search all relevant electronic databases up to 2006. The authors also hand-searched the reference lists of key articles and reviews. They included randomised controlled trials (RCT) involving patients in primary care who were not seeking alcohol treatment and who received brief intervention. Two authors independently abstracted data and assessed trial quality. Random effects meta-analyses, subgroup and sensitivity analyses and meta-regression were conducted.This resource was contributed by The National Documentation Centre on Drug Use.

Semi-online neural-Q_leaming for real-time robot learning

Reinforcement learning (RL) is a very suitable technique for robot learning, as it can learn in unknown environments and in real-time computation. The main difficulties in adapting classic RL algorithms to robotic systems are the generalization problem and the correct observation of the Markovian state. This paper attempts to solve the generalization problem by proposing the semi-online neural-Q_learning algorithm (SONQL). The algorithm uses the classic Q_learning technique with two modifications. First, a neural network (NN) approximates the Q_function allowing the use of continuous states and actions. Second, a database of the most representative learning samples accelerates and stabilizes the convergence. The term semi-online is referred to the fact that the algorithm uses the current but also past learning samples. However, the algorithm is able to learn in real-time while the robot is interacting with the environment. The paper shows simulated results with the "mountain-car" benchmark and, also, real results with an underwater robot in a target following behavior

Uncertainty Quantification Of A Semi-Supervised Support Vector Regression Reservoir Model

Uncertainty quantification of petroleum reservoir models is one of the present challenges, which is usually approached with a wide range of geostatistical tools linked with statistical optimisation or/and inference algorithms. Recent advances in machine learning offer a novel approach to model spatial distribution of petrophysical properties in complex reservoirs alternative to geostatistics. The approach is based of semisupervised learning, which handles both ?labelled? observed data and ?unlabelled? data, which have no measured value but describe prior knowledge and other relevant data in forms of manifolds in the input space where the modelled property is continuous. Proposed semi-supervised Support Vector Regression (SVR) model has demonstrated its capability to represent realistic geological features and describe stochastic variability and non-uniqueness of spatial properties. On the other hand, it is able to capture and preserve key spatial dependencies such as connectivity of high permeability geo-bodies, which is often difficult in contemporary petroleum reservoir studies. Semi-supervised SVR as a data driven algorithm is designed to integrate various kind of conditioning information and learn dependences from it. The semi-supervised SVR model is able to balance signal/noise levels and control the prior belief in available data. In this work, stochastic semi-supervised SVR geomodel is integrated into Bayesian framework to quantify uncertainty of reservoir production with multiple models fitted to past dynamic observations (production history). Multiple history matched models are obtained using stochastic sampling and/or MCMC-based inference algorithms, which evaluate posterior probability distribution. Uncertainty of the model is described by posterior probability of the model parameters that represent key geological properties: spatial correlation size, continuity strength, smoothness/variability of spatial property distribution. The developed approach is illustrated with a fluvial reservoir case. The resulting probabilistic production forecasts are described by uncertainty envelopes. The paper compares the performance of the models with different combinations of unknown parameters and discusses sensitivity issues.

Embedding Emotional Context in Recommender Systems

Emotions are crucial for user's decision making in recommendation processes. We first introduce ambient recommender systems, which arise from the analysis of new trends on the exploitation of the emotional context in the next generation of recommender systems. We then explain some results of these new trends in real-world applications through the smart prediction assistant (SPA) platform in an intelligent learning guide with more than three million users. While most approaches to recommending have focused on algorithm performance. SPA makes recommendations to users on the basis of emotional information acquired in an incremental way. This article provides a cross-disciplinary perspective to achieve this goal in such recommender systems through a SPA platform. The methodology applied in SPA is the result of a bunch of technology transfer projects for large real-world rccommender systems

The appropriateness of renal angioplasty: the ANPARIA software: a multidisciplinary expert panel approach

Percutaneous transluminal renal angioplasty (PTRA) is an invasive technique that is costly and involves the risk of complications and renal failure. The ability of PTRA to reduce the administration of antihypertensive drugs has been demonstrated. A potentially greater benefit, which nevertheless remains to be proven, is the deferral of the need for chronic dialysis. The aim of the study (ANPARIA) was to assess the appropriateness of PTRA to impact on the evolution of renal function. A standardized expert panel method was used to assess the appropriateness of medical treatment alone or medical treatment with revascularization in various clinical situations. The choice of revascularization by either PTRA or surgery was examined for each clinical situation. Analysis was based on a detailed literature review and on systematically elicited expert opinion, which were obtained during a two-round modified Delphi process. The study provides detailed responses on the appropriateness of PTRA for 1848 distinct clinical scenarios. Depending on the major clinical presentation, appropriateness of revascularization varied from 32% to 75% for individual scenarios (overal 48%). Uncertainty as to revascularization was 41% overall. When revascularization was appropriate, PTRA was favored over surgery in 94% of the scenarios, except in certain cases of aortic atheroma where sugery was the preferred choice. Kidney size [7 cm, absence of coexisting disease, acute renal failure, a high degree of stenosis (C70%), and absence of multiple arteries were identified as predictive variables of favorable appropriateness ratings. Situations such as cardiac failure with pulmonary edema or acute thrombosis of the renal artery were defined as indications for PTRA. This study identified clinical situations in which PTRA or surgery are appropriate for renal artery disease. We built a decision tree which can be used via Internet: the ANPARIA software (http://www.chu-clermontferrand.fr/anparia/). In numerous clinical situations uncertainty remains as to whether PTRA prevents deterioration of renal function.

Acute Effects of Cannabis Smoking on Skills Related to Driving : an fMRI Study

Introduction : Driving is a complex everyday task requiring mechanisms of perception, attention, learning, memory, decision making and action control, thus indicating that involves numerous and varied brain networks. If many data have been accumulated over time about the effects of alcohol consumption on driving capability, much less is known about the role of other psychoactive substances, such as cannabis (Chang et al.2007, Ramaekers et al, 2006). Indeed, the solicited brain areas during safe driving which could be affected by cannabis exposure have not yet been clearly identified. Our aim is to study these brain regions during a tracking task related to driving skills and to evaluate the modulation due to the tolerance of cannabis effects. Methods : Eight non-smoker control subjects participated to an fMRI experiment based on a visuo-motor tracking task, alternating active tracking blocks with passive tracking viewing and rest condition. Half of the active tracking conditions included randomly presented traffic lights as distractors. Subjects were asked to track with a joystick with their right hand and to press a button with their left index at each appearance of a distractor. Four smoking subjects participated to the same fMRI sessions once before and once after smoking cannabis and a placebo in two independent cross-over experiments. We quantified the performance of the subjects by measuring the precision of the behavioural responses (i.e. percentage of time of correct tracking and reaction times to distractors). Functional MRI data were acquired using on a 3.0T Siemens Trio system equipped with a 32-channel head coil. BOLD signals will be obtained with a gradient-echo EPI sequence (TR=2s, TE=30ms, FoV=216mm, FA=90°, matrix size 72×72, 32 slices, thickness 3mm). Preprocessing, single subject analysis and group statistics were conducted on SPM8b. Results were thresholded at p<0.05 (FWE corrected) and at k>30 for spatial extent. Results : Behavioural results showed a significant impairment in task and cognitive test performance of the subjects after cannabis inhalation when comparing their tracking accuracy either to the controls subjects or to their performances before the inhalation or after the placebo inhalation (p<0.001 corrected). In controls, fMRI BOLD analysis of the active tracking condition compared to the passive one revealed networks of polymodal areas in superior frontal and parietal cortex dealing with attention and visuo-spatial coordination. In accordance to what is known of the visual and sensory motor networks we found activations in V4, frontal eye-field, right middle frontal gyrus, intra-parietal sulcus, temporo-parietal junction, premotor and sensory-motor cortex. The presence of distractors added a significant activation in the precuneus. Preliminary results on cannabis smokers in the acute phase, compared either to themselves before the cannabis inhalation or to control subjects, showed a decreased activation in large portions of the frontal and parietal attention network during the simple tracking task, but greater involvement of precuneus, of the superior part of intraparietal sulcus and middle frontal gyrus bilaterally when distractors were present in the task. Conclusions : Our preliminary results suggest that acute cannabis smoking alters performances and brain activity during active tracking tasks, partly reorganizing the recruitment of brain areas of the attention network.

Learning spatial filters for multispectral image segmentation

We present a novel filtering method for multispectral satellite image classification. The proposed method learns a set of spatial filters that maximize class separability of binary support vector machine (SVM) through a gradient descent approach. Regularization issues are discussed in detail and a Frobenius-norm regularization is proposed to efficiently exclude uninformative filters coefficients. Experiments carried out on multiclass one-against-all classification and target detection show the capabilities of the learned spatial filters.