Active Learning of Very-High Resolution Optical Imagery with SVM: Entropy vs Margin Sampling

An active learning method is proposed for the semi-automatic selection of training sets in remote sensing image classification. The method adds iteratively to the current training set the unlabeled pixels for which the prediction of an ensemble of classifiers based on bagged training sets show maximum entropy. This way, the algorithm selects the pixels that are the most uncertain and that will improve the model if added in the training set. The user is asked to label such pixels at each iteration. Experiments using support vector machines (SVM) on an 8 classes QuickBird image show the excellent performances of the methods, that equals accuracies of both a model trained with ten times more pixels and a model whose training set has been built using a state-of-the-art SVM specific active learning method

Missing value imputation in longitudinal measures of alcohol consumption

Attrition in longitudinal studies can lead to biased results. The study is motivated by the unexpected observation that alcohol consumption decreased despite increased availability, which may be due to sample attrition of heavy drinkers. Several imputation methods have been proposed, but rarely compared in longitudinal studies of alcohol consumption. The imputation of consumption level measurements is computationally particularly challenging due to alcohol consumption being a semi-continuous variable (dichotomous drinking status and continuous volume among drinkers), and the non-normality of data in the continuous part. Data come from a longitudinal study in Denmark with four waves (2003-2006) and 1771 individuals at baseline. Five techniques for missing data are compared: Last value carried forward (LVCF) was used as a single, and Hotdeck, Heckman modelling, multivariate imputation by chained equations (MICE), and a Bayesian approach as multiple imputation methods. Predictive mean matching was used to account for non-normality, where instead of imputing regression estimates, "real" observed values from similar cases are imputed. Methods were also compared by means of a simulated dataset. The simulation showed that the Bayesian approach yielded the most unbiased estimates for imputation. The finding of no increase in consumption levels despite a higher availability remained unaltered. Copyright (C) 2011 John Wiley & Sons, Ltd.

Not in Education, Employment, or Training status among young Swiss men. Longitudinal associations with mental health and substance use.

PURPOSE: Not in Education, Employment, or Training (NEET) youth are youth disengaged from major social institutions and constitute a worrying concern. However, little is known about this subgroup of vulnerable youth. This study aimed to examine if NEET youth differ from other contemporaries in terms of personality, mental health, and substance use and to provide longitudinal examination of NEET status, testing its stability and prospective pathways with mental health and substance use. METHODS: As part of the Cohort Study on Substance Use Risk Factors, 4,758 young Swiss men in their early 20s answered questions concerning their current professional and educational status, personality, substance use, and symptomatology related to mental health. Descriptive statistics, generalized linear models for cross-sectional comparisons, and cross-lagged panel models for longitudinal associations were computed. RESULTS: NEET youth were 6.1% at baseline and 7.4% at follow-up with 1.4% being NEET at both time points. Comparisons between NEET and non-NEET youth showed significant differences in substance use and depressive symptoms only. Longitudinal associations showed that previous mental health, cannabis use, and daily smoking increased the likelihood of being NEET. Reverse causal paths were nonsignificant. CONCLUSIONS: NEET status seemed to be unlikely and transient among young Swiss men, associated with differences in mental health and substance use but not in personality. Causal paths presented NEET status as a consequence of mental health and substance use rather than a cause. Additionally, this study confirmed that cannabis use and daily smoking are public health problems. Prevention programs need to focus on these vulnerable youth to avoid them being disengaged.

The impact of social relationships on the maintenance of independence in advanced old age: findings of a Swiss Longitudinal Study .

The impact of social relationships on the maintenance of independence over periods of 12-18 months in a group of 306 octogenarians is assessed in this study. The study is based on the results of the Swilsoo (Swiss Interdisciplinary Longitudinal Study on the Oldest Old). Participants (80-84 years old at baseline) were interviewed five times between 1994 and 1999. Independence was defined as the capacity to perform without assistance eight activities of daily living. We distinguished in our analyses kinship and friendship networks and evaluated social relationships with the help of a series of variables serving as indicators of network composition and contact frequency. Logistic regression models were used to identify the short-term effects of social relationships on independence, after controlling for sociodemographic and health-related variables; independence at a given wave of interviews was interpreted in the light of social factors measured at the previous wave. Our analyses indicate that the existence of a close friend has a significant impact on the maintenance of independence (OR=1.58, p<0.05), which is not the case with the other variables concerning network composition. Kinship contacts were also observed to have a positive impact on independence (OR=1.12, p<0.01).

Effect of a national primary care pay for performance scheme on emergency hospital admissions for ambulatory care sensitive conditions: controlled longitudinal study.

OBJECTIVE: To estimate the impact of a national primary care pay for performance scheme, the Quality and Outcomes Framework in England, on emergency hospital admissions for ambulatory care sensitive conditions (ACSCs). DESIGN: Controlled longitudinal study. SETTING: English National Health Service between 1998/99 and 2010/11. PARTICIPANTS: Populations registered with each of 6975 family practices in England. MAIN OUTCOME MEASURES: Year specific differences between trend adjusted emergency hospital admission rates for incentivised ACSCs before and after the introduction of the Quality and Outcomes Framework scheme and two comparators: non-incentivised ACSCs and non-ACSCs. RESULTS: Incentivised ACSC admissions showed a relative reduction of 2.7% (95% confidence interval 1.6% to 3.8%) in the first year of the Quality and Outcomes Framework compared with ACSCs that were not incentivised. This increased to a relative reduction of 8.0% (6.9% to 9.1%) in 2010/11. Compared with conditions that are not regarded as being influenced by the quality of ambulatory care (non-ACSCs), incentivised ACSCs also showed a relative reduction in rates of emergency admissions of 2.8% (2.0% to 3.6%) in the first year increasing to 10.9% (10.1% to 11.7%) by 2010/11. CONCLUSIONS: The introduction of a major national pay for performance scheme for primary care in England was associated with a decrease in emergency admissions for incentivised conditions compared with conditions that were not incentivised. Contemporaneous health service changes seem unlikely to have caused the sharp change in the trajectory of incentivised ACSC admissions immediately after the introduction of the Quality and Outcomes Framework. The decrease seems larger than would be expected from the changes in the process measures that were incentivised, suggesting that the pay for performance scheme may have had impacts on quality of care beyond the directly incentivised activities.

Complete longitudinal analyses of the randomized, placebo-controlled, phase III trial of sunitinib in patients with gastrointestinal stromal tumor following imatinib failure.

PURPOSE: To analyze final long-term survival and clinical outcomes from the randomized phase III study of sunitinib in gastrointestinal stromal tumor patients after imatinib failure; to assess correlative angiogenesis biomarkers with patient outcomes. EXPERIMENTAL DESIGN: Blinded sunitinib or placebo was given daily on a 4-week-on/2-week-off treatment schedule. Placebo-assigned patients could cross over to sunitinib at disease progression/study unblinding. Overall survival (OS) was analyzed using conventional statistical methods and the rank-preserving structural failure time (RPSFT) method to explore cross-over impact. Circulating levels of angiogenesis biomarkers were analyzed. RESULTS: In total, 243 patients were randomized to receive sunitinib and 118 to placebo, 103 of whom crossed over to open-label sunitinib. Conventional statistical analysis showed that OS converged in the sunitinib and placebo arms (median 72.7 vs. 64.9 weeks; HR, 0.876; P = 0.306) as expected, given the cross-over design. RPSFT analysis estimated median OS for placebo of 39.0 weeks (HR, 0.505, 95% CI, 0.262-1.134; P = 0.306). No new safety concerns emerged with extended sunitinib treatment. No consistent associations were found between the pharmacodynamics of angiogenesis-related plasma proteins during sunitinib treatment and clinical outcome. CONCLUSIONS: The cross-over design provided evidence of sunitinib clinical benefit based on prolonged time to tumor progression during the double-blind phase of this trial. As expected, following cross-over, there was no statistical difference in OS. RPSFT analysis modeled the absence of cross-over, estimating a substantial sunitinib OS benefit relative to placebo. Long-term sunitinib treatment was tolerated without new adverse events.

Impact of vaccines given during pregnancy on the offspring of women consulting a travel clinic: a longitudinal study

BACKGROUND: Little is known on the impact of travel vaccinations during pregnancy on child outcomes, in particular on the long-term psychomotor development. The objectives of the study were (1) to estimate the rate of premature births, congenital abnormalities, and mental and physical development problems of children born from mothers who had been vaccinated during pregnancy and (2) to compare these rates with those of children whose mothers had not been vaccinated during pregnancy. METHODS: Longitudinal study including (1) retrospectively pregnant women having attended our travel clinic before (vaccinated) and (2) prospectively mothers attending our clinic (nonvaccinated). We performed phone interviews with mothers vaccinated during pregnancy, up to 10 years before, and face-to-face interviews with nonvaccinated age-matched mothers, ie, women attending the travel clinic who had one child of about the same age as the one of the case to compare child development between both groups. RESULTS: Fifty-three women vaccinated during pregnancy were interviewed as well as 53 nonvaccinated ones. Twenty-eight (53%) women received their vaccination during the first trimester. The most frequent vaccine administered was hepatitis A (55% of the cases), followed by di-Te (34%), IM poliomyelitis (23%), yellow fever (12%), A-C meningitis (8%), IM typhoid (4%), and oral poliomyelitis (4%). Children were followed for a range of 1 to 10 years. Rates of premature births were 5.7% in both groups; congenital abnormalities were 1.9% in the vaccinated cohort versus 5.7% in the nonvaccinated one; children took their first steps at a median age of 12 months in both cohorts; among schoolchildren, 5% of the vaccinated cohort versus 7.7% of the nonvaccinated attended a lower level or a specialized school. CONCLUSION: In this small sample size, there was no indication that usual travel vaccinations, including the yellow fever one, had deleterious effect on child outcome and development

Longitudinal analysis of patterns and predictors of changes in self-reported adherence to antiretroviral therapy: Swiss HIV Cohort Study.

BACKGROUND: Adherence to combination antiretroviral therapy (cART) is a dynamic process, however, changes in adherence behavior over time are insufficiently understood. METHODS: Data on self-reported missed doses of cART was collected every 6 months in Swiss HIV Cohort Study participants. We identified behavioral groups associated with specific cART adherence patterns using trajectory analyses. Repeated measures logistic regression identified predictors of changes in adherence between consecutive visits. RESULTS: Six thousand seven hundred nine individuals completed 49,071 adherence questionnaires [median 8 (interquartile range: 5-10)] during a median follow-up time of 4.5 years (interquartile range: 2.4-5.1). Individuals were clustered into 4 adherence groups: good (51.8%), worsening (17.4%), improving (17.6%), and poor adherence (13.2%). Independent predictors of worsening adherence were younger age, basic education, loss of a roommate, starting intravenous drug use, increasing alcohol intake, depression, longer time with HIV, onset of lipodystrophy, and changing care provider. Independent predictors of improvements in adherence were regimen simplification, changing class of cART, less time on cART, and starting comedications. CONCLUSIONS: Treatment, behavioral changes, and life events influence patterns of drug intake in HIV patients. Clinical care providers should routinely monitor factors related to worsening adherence and intervene early to reduce the risk of treatment failure and drug resistance.

Predictive ability of an early diagnostic guess in patients presenting with chest pain; a longitudinal descriptive study.

ABSTRACT: BACKGROUND: The intuitive early diagnostic guess could play an important role in reaching a final diagnosis. However, no study to date has attempted to quantify the importance of general practitioners' (GPs) ability to correctly appraise the origin of chest pain within the first minutes of an encounter. METHODS: The validation study was nested in a multicentre cohort study with a one year follow-up and included 626 successive patients who presented with chest pain and were attended by 58 GPs in Western Switzerland. The early diagnostic guess was assessed prior to a patient's history being taken by a GP and was then compared to a diagnosis of chest pain observed over the next year. RESULTS: Using summary measures clustered at the GP's level, the early diagnostic guess was confirmed by further investigation in 51.0% (CI 95%; 49.4% to 52.5%) of patients presenting with chest pain. The early diagnostic guess was more accurate in patients with a life threatening illness (65.4%; CI 95% 64.5% to 66.3%) and in patients who did not feel anxious (62.9%; CI 95% 62.5% to 63.3%). The predictive abilities of an early diagnostic guess were consistent among GPs. CONCLUSIONS: The GPs early diagnostic guess was correct in one out of two patients presenting with chest pain. The probability of a correct guess was higher in patients with a life-threatening illness and in patients not feeling anxious about their pain.

Quantitative ultrasound of bone in institutionalized elderly women: a cross-sectional and longitudinal study.

Quantitative ultrasound of bone is a promising method for bone assessment: radiation-free, portable and predictive of hip fracture. Its portability allowed us to study the relationships between ultrasonic parameters of bone with age and with non-vertebral fractures in elderly women living in 19 nursing homes. Broadband ultrasound attenuation (BUA) and speed of sound (SOS) of the calcaneus were measured (and the stiffness index calculated) in a sample of 270 institutionalized women, aged 85 +/- 7 years, using an Achilles bone densitometer (Lunar). The effects of age, history of non-vertebral and non-traumatic fractures, body mass index, triceps skinfold and arm circumference were assessed on BUA, SOS and stiffness index. Furthermore, to evaluate longitudinally the influence of aging on the ultrasound parameters of bone, 60 subjects from the same group had a second ultrasound measurement after 1 year. The cross-sectional analysis of the data on all 270 women showed a significant decrease (p < 0.001) with age in BUA, SOS and stiffness index (-0.47%, -0.06%, and -1.01% respectively per year). In the 94 women, (35%) with a history of previous non-vertebral fractures, ultrasound parameters were significantly lower (p < 0.0001) than in the 176 women with no history of fracture (-8.3% for BUA, -1.3% for SOS, -18.9% for stiffness index). In contrast, there was no significant difference in anthropometric measurements between the groups with and without previous non-vertebral fractures, although the measurements decreased significantly with age. In the longitudinal study, repeated quantitative ultrasound after 11.4 +/- 0.8 months showed no significant decrease in BUA (-1%) but a significant decrease in SOS (-0.3%, p < 0.0001) and in stiffness index (-3.6%, p < 0.0002). In conclusion, quantitative ultrasound of the calcaneus measures properties of bone which continue to decline in institutionalized elderly women, and is able to discriminate women with previous non-vertebral fractures.

Increased audiovisual integration in cochlear-implanted deaf patients: independent components analysis of longitudinal positron emission tomography data.

It has been demonstrated in earlier studies that patients with a cochlear implant have increased abilities for audio-visual integration because the crude information transmitted by the cochlear implant requires the persistent use of the complementary speech information from the visual channel. The brain network for these abilities needs to be clarified. We used an independent components analysis (ICA) of the activation (H2 (15) O) positron emission tomography data to explore occipito-temporal brain activity in post-lingually deaf patients with unilaterally implanted cochlear implants at several months post-implantation (T1), shortly after implantation (T0) and in normal hearing controls. In between-group analysis, patients at T1 had greater blood flow in the left middle temporal cortex as compared with T0 and normal hearing controls. In within-group analysis, patients at T0 had a task-related ICA component in the visual cortex, and patients at T1 had one task-related ICA component in the left middle temporal cortex and the other in the visual cortex. The time courses of temporal and visual activities during the positron emission tomography examination at T1 were highly correlated, meaning that synchronized integrative activity occurred. The greater involvement of the visual cortex and its close coupling with the temporal cortex at T1 confirm the importance of audio-visual integration in more experienced cochlear implant subjects at the cortical level.