Fully automatic segmentation of AP pelvis X-rays via random forest regression with efficient feature selection and hierarchical sparse shape composition

In clinical practice, traditional X-ray radiography is widely used, and knowledge of landmarks and contours in anteroposterior (AP) pelvis X-rays is invaluable for computer aided diagnosis, hip surgery planning and image-guided interventions. This paper presents a fully automatic approach for landmark detection and shape segmentation of both pelvis and femur in conventional AP X-ray images. Our approach is based on the framework of landmark detection via Random Forest (RF) regression and shape regularization via hierarchical sparse shape composition. We propose a visual feature FL-HoG (Flexible- Level Histogram of Oriented Gradients) and a feature selection algorithm based on trace radio optimization to improve the robustness and the efficacy of RF-based landmark detection. The landmark detection result is then used in a hierarchical sparse shape composition framework for shape regularization. Finally, the extracted shape contour is fine-tuned by a post-processing step based on low level image features. The experimental results demonstrate that our feature selection algorithm reduces the feature dimension in a factor of 40 and improves both training and test efficiency. Further experiments conducted on 436 clinical AP pelvis X-rays show that our approach achieves an average point-to-curve error around 1.2 mm for femur and 1.9 mm for pelvis.

On Graphically Checking Goodness-of-fit of Binary Logistic Regression Models

OBJECTIVES: This paper is concerned with checking goodness-of-fit of binary logistic regression models. For the practitioners of data analysis, the broad classes of procedures for checking goodness-of-fit available in the literature are described. The challenges of model checking in the context of binary logistic regression are reviewed. As a viable solution, a simple graphical procedure for checking goodness-of-fit is proposed. METHODS: The graphical procedure proposed relies on pieces of information available from any logistic analysis; the focus is on combining and presenting these in an informative way. RESULTS: The information gained using this approach is presented with three examples. In the discussion, the proposed method is put into context and compared with other graphical procedures for checking goodness-of-fit of binary logistic models available in the literature. CONCLUSION: A simple graphical method can significantly improve the understanding of any logistic regression analysis and help to prevent faulty conclusions.

Direct adipotropic actions of atorvastatin: differentiation state-dependent induction of apoptosis, modulation of endocrine function, and inhibition of glucose uptake

Statins exert anti-inflammatory, anti-atherogenic actions. The mechanisms responsible for these effects remain only partially elucidated. Diabetes and obesity are characterized by low-grade inflammation. Metabolic and endocrine adipocyte dysfunction is known to play a crucial role in the development of these disorders and the related cardiovascular complications. Thus, direct modulation of adipocyte function may represent a mechanism of pleiotropic statin actions. We investigated effects of atorvastatin on apoptosis, differentiation, endocrine, and metabolic functions in murine white and brown adipocyte lines. Direct exposure of differentiating preadipocytes to atorvastatin strongly reduced lipid accumulation and diminished protein expression of the differentiation marker CCAAT/enhancer binding protein-beta (CEBP-beta). In fully differentiated adipocytes, however, lipid accumulation remained unchanged after chronic atorvastatin treatment. Furthermore, cell viability was reduced in response to atorvastatin treatment in proliferating and differentiating preadipocytes, but not in differentiated cells. Moreover, atorvastatin induced apoptosis and inhibited protein kinase B (AKT) phosphorylation in proliferating and differentiating preadipocytes, but not in differentiated adipocytes. On the endocrine level, direct atorvastatin treatment of differentiated white adipocytes enhanced expression of the pro-inflammatory adipokine interleukin-6 (IL-6), and downregulated expression of the insulin-mimetic and anti-inflammatory adipokines visfatin and adiponectin. Finally, these direct adipotropic endocrine effects of atorvastatin were paralleled by the acute inhibition of insulin-induced glucose uptake in differentiated white adipocytes, while protein expression of the thermogenic uncoupling protein-1 (UCP-1) in brown adipocytes remained unchanged. Taken together, our data for the first time demonstrate direct differentiation state-dependent effects of atorvastatin including apoptosis, modulation of pro-inflammatory and glucostatic adipokine expression, and insulin resistance in adipose cells. These differential interactions may explain variable clinical observations.

Frequency-domain source localization shows state-dependent diazepam effects in 47-channel EEG

The topic of this study was to evaluate state-dependent effects of diazepam on the frequency characteristics of 47-channel spontaneous EEG maps. A novel method, the FFT-Dipole-Approximation (Lehmann and Michel, 1990), was used to study effects on the strength and the topography of the maps in the different frequency bands. Map topography was characterized by the 3-dimensional location of the equivalent dipole source and map strength was defined as the spatial standard deviation (the Global Field Power) of the maps of each frequency point. The Global Field Power can be considered as a measure of the amount of energy produced by the system, while the source location gives an estimate of the center of gravity of all sources in the brain that were active at a certain frequency. State-dependency was studied by evaluating the drug effects before and after a continuous performance task of 25 min duration. Clear interactions between drug (diazepam vs. placebo) and time after drug intake (before and after the task) were found, especially in the inferior-superior location of the dipole sources. It supports the hypothesis that diazepam, like other drugs, has different effects on brain functions depending on the momentary functional state of the brain. In addition to the drug effects, clearly different source locations and Global Field Power were found for the different frequency bands, replicating earlier reports (Michel et al., 1992).

Potentially avoidable 30-day hospital readmissions in medical patients: derivation and validation of a prediction model

IMPORTANCE Because effective interventions to reduce hospital readmissions are often expensive to implement, a score to predict potentially avoidable readmissions may help target the patients most likely to benefit. OBJECTIVE To derive and internally validate a prediction model for potentially avoidable 30-day hospital readmissions in medical patients using administrative and clinical data readily available prior to discharge. DESIGN Retrospective cohort study. SETTING Academic medical center in Boston, Massachusetts. PARTICIPANTS All patient discharges from any medical services between July 1, 2009, and June 30, 2010. MAIN OUTCOME MEASURES Potentially avoidable 30-day readmissions to 3 hospitals of the Partners HealthCare network were identified using a validated computerized algorithm based on administrative data (SQLape). A simple score was developed using multivariable logistic regression, with two-thirds of the sample randomly selected as the derivation cohort and one-third as the validation cohort. RESULTS Among 10 731 eligible discharges, 2398 discharges (22.3%) were followed by a 30-day readmission, of which 879 (8.5% of all discharges) were identified as potentially avoidable. The prediction score identified 7 independent factors, referred to as the HOSPITAL score: h emoglobin at discharge, discharge from an o ncology service, s odium level at discharge, p rocedure during the index admission, i ndex t ype of admission, number of a dmissions during the last 12 months, and l ength of stay. In the validation set, 26.7% of the patients were classified as high risk, with an estimated potentially avoidable readmission risk of 18.0% (observed, 18.2%). The HOSPITAL score had fair discriminatory power (C statistic, 0.71) and had good calibration. CONCLUSIONS AND RELEVANCE This simple prediction model identifies before discharge the risk of potentially avoidable 30-day readmission in medical patients. This score has potential to easily identify patients who may need more intensive transitional care interventions.

Predicting smoking cessation and its relapse in HIV-infected patients: the Swiss HIV Cohort Study

OBJECTIVES: The aim of the study was to assess whether prospective follow-up data within the Swiss HIV Cohort Study can be used to predict patients who stop smoking; or among smokers who stop, those who start smoking again. METHODS: We built prediction models first using clinical reasoning ('clinical models') and then by selecting from numerous candidate predictors using advanced statistical methods ('statistical models'). Our clinical models were based on literature that suggests that motivation drives smoking cessation, while dependence drives relapse in those attempting to stop. Our statistical models were based on automatic variable selection using additive logistic regression with component-wise gradient boosting. RESULTS: Of 4833 smokers, 26% stopped smoking, at least temporarily; because among those who stopped, 48% started smoking again. The predictive performance of our clinical and statistical models was modest. A basic clinical model for cessation, with patients classified into three motivational groups, was nearly as discriminatory as a constrained statistical model with just the most important predictors (the ratio of nonsmoking visits to total visits, alcohol or drug dependence, psychiatric comorbidities, recent hospitalization and age). A basic clinical model for relapse, based on the maximum number of cigarettes per day prior to stopping, was not as discriminatory as a constrained statistical model with just the ratio of nonsmoking visits to total visits. CONCLUSIONS: Predicting smoking cessation and relapse is difficult, so that simple models are nearly as discriminatory as complex ones. Patients with a history of attempting to stop and those known to have stopped recently are the best candidates for an intervention.

Self-esteem development across the life span: A longitudinal study with a large sample from Germany

The authors examined the development of self-esteem across the life span. Data came from a German longitudinal study with 3 assessments across 4 years of a sample of 2,509 individuals ages 14 to 89 years. The self-esteem measure used showed strong measurement invariance across assessments and birth cohorts. Latent growth curve analyses indicated that self-esteem follows a quadratic trajectory across the life span, increasing during adolescence, young adulthood, and middle adulthood, reaching a peak at age 60 years, and then declining in old age. No cohort effects on average levels of self-esteem or on the shape of the trajectory were found. Moreover, the trajectory did not differ across gender, level of education, or for individuals who had lived continuously in West versus East Germany (i.e., the 2 parts of Germany that had been separate states from 1949 to 1990). However, the results suggested that employment status, household income, and satisfaction in the domains of work, relationships, and health contribute to a more positive life span trajectory of self-esteem. The findings have significant implications, because they call attention to developmental stages in which individuals may be vulnerable because of low self-esteem (such as adolescence and old age) and to factors that predict successful versus problematic developmental trajectories.

Consistency of Concave Regression with an Application to Current-Status Data

We consider the problem of nonparametric estimation of a concave regression function F. We show that the supremum distance between the least square s estimatorand F on a compact interval is typically of order(log(n)/n)2/5. This entails rates of convergence for the estimator’s derivative. Moreover, we discuss the impact of additional constraints on F such as monotonicity and pointwise bounds. Then we apply these results to the analysis of current status data, where the distribution function of the event times is assumed to be concave.

Invited Abstract: Issues with State-based Energy Consumption Modelling

Energy consumption modelling by state based approaches often assume constant energy consumption values in each state. However, it happens in certain situations that during state transitions or even during a state the energy consumption is not constant and does fluctuate. This paper discusses those issues by presenting some examples from wireless sensor and wireless local area networks for such cases and possible solutions.

Effects of dialysis modality on blood loss, bleeding complications and transfusion requirements in critically ill patients with dialysis-dependent acute renal failure.

Blood loss and bleeding complications may often be observed in critically ill patients on renal replacement therapies (RRT). Here we investigate procedural (i.e. RRT-related) and non-procedural blood loss as well as transfusion requirements in regard to the chosen mode of dialysis (i.e. intermittent haemodialysis [IHD] versus continuous veno-venous haemofiltration [CVVH]). Two hundred and fifty-two patients (122 CVVH, 159 male; aged 61.5±13.9 years) with dialysis-dependent acute renal failure were analysed in a sub-analysis of the prospective randomised controlled clinical trial-CONVINT-comparing IHD and CVVH. Bleeding complications including severity of bleeding and RRT-related blood loss were assessed. We observed that 3.6% of patients died related to severe bleeding episodes (between group P=0.94). Major all-cause bleeding complications were observed in 23% IHD versus 26% of CVVH group patients (P=0.95). Under CVVH, the rate of RRT-related blood loss events (57.4% versus 30.4%, P=0.01) and mean total blood volume lost was increased (222.3±291.9 versus 112.5±222.7 ml per patient, P <0.001). Overall, transfusion rates did not differ between the study groups. In patients with sepsis, transfusion rates of all blood products were significantly higher when compared to cardiogenic shock (all P <0.01) or other conditions. In conclusion, procedural and non-procedural blood loss may often be observed in critically ill patients on RRT. In CVVH-treated patients, procedural blood loss was increased but overall transfusion rates remained unchanged. Our data show that IHD and CVVH may be regarded as equivalent approaches in critically ill patients with dialysis-dependent acute renal failure in this regard.

LOGITCPRPLOT: Stata module to graph component-plus-residual plot for logistic regression

logitcprplot can be used after logistic regression for graphing a component-plus-residual plot (a.k.a. partial residual plot) for a given predictor, including a lowess, local polynomial, restricted cubic spline, fractional polynomial, penalized spline, regression spline, running line, or adaptive variable span running line smooth

RRLOGIT: Stata module to estimate logistic regression for randomized response data

rrlogit fits a maximum-likelihood logistic regression for randomized response data.