Prediction of conversion from mild cognitive impairment to Alzheimer disease based on bayesian data mining with ensemble learning

Prediction of disease progress is of great importance to Alzheimer disease (AD) researchers and clinicians. Previous attempts at constructing predictive models have been hindered by undersampling, and restriction to linear associations among variables, among other problems. To address these problems, we propose a novel Bayesian data-mining method called Bayesian Outcome Prediction with Ensemble Learning (BOPEL). BOPEL uses a Bayesian-network representation with boosting, to allow the detection of nonlinear multivariate associations, and incorporates resampling-based feature selection to prevent over-fitting caused by undersampling. We demonstrate the use of this approach in predicting conversion to AD in individuals with mild cognitive impairment (MCI), based on structural magnetic-resonance and magnetic-resonance- spectroscopy data. This study includes 26 subjects with amnestic MCI: the converter group (n = 8) met MCI criteria at baseline, but converted to AD within five years, whereas the non-converter group (n = 18) met MCI criteria at baseline and at follow-up. We found that BOPEL accurately differentiates MCI converters from non-converters, based on the baseline volumes of the left hippocampus, the banks of the right superior temporal sulcus, the right entorhinal cortex, the left lingual gyrus, and the rostral aspect of the left middle frontal gyrus. Prediction accuracy was 0.81, sensitivity was 0.63 and specificity was 0.89. We validated the generated predictive model with an independent data set constructed from the Alzheimer Disease Neuroimaging Initiative database, and again found high predictive accuracy (0.75).

Perceived Muscle Soreness in Recreational Female Runners

The purpose of this study was to determine if rating of perceived exertion correlated with perceived muscle soreness during delayed onset muscle soreness (DOMS) in female runners. This study examined the pre and post running economy measures and perceived muscle soreness before and after a 30-min downhill run (DHR) at -15% grade and 70% of the subjects predetermined maximum oxygen uptake (VO2 peak). Six female recreational runners (mean age = 24.5) performed level running at 65%, 75%, and 85% of their VO2 peak prior to DHR (baseline economy runs), as well as, immediately following and 4 successive days after the DHR.

RESULTS

Subjective response related to perceived muscle soreness increased significantly from a mean of 2 (pre DHR) to 62 (2 days post DHR) on a scale of 1-100. Creatine kinase levels and oxygen consumption increased post DHR compared to pre DHR. Rating of perceived exertion did not change between the economy runs performed prior to or at any point after the DHR.

CONCLUSION

Perceived muscle soreness is a better tool than the RPE scale to monitor exercise intensity for recreational female runners during periods of DOMS and running economy is adversely affected by DOMS.

Network-level analysis of cortical thickness of the epileptic brain

Temporal lobe epilepsy (TLE) characterized by an epileptogenic focus in the medial temporal lobe is the most common form of focal epilepsy. However, the seizures are not confined to the temporal lobe but can spread to other, anatomically connected brain regions where they can cause similar structural abnormalities as observed in the focus. The aim of this study was to derive whole-brain networks from volumetric data and obtain network-centric measures, which can capture cortical thinning characteristic of TLE and can be used for classifying a given MRI into TLE or normal, and to obtain additional summary statistics that relate to the extent and spread of the disease. T1-weighted whole-brain images were acquired on a 4-T magnet in 13 patients with TLE with mesial temporal lobe sclerosis (TLE-MTS), 14 patients with TLE with normal MRI (TLE-no), and 30 controls. Mean cortical thickness and curvature measurements were obtained using the FreeSurfer software. These values were used to derive a graph, or network, for each subject. The nodes of the graph are brain regions, and edges represent disease progression paths. We show how to obtain summary statistics like mean, median, and variance defined for these networks and to perform exploratory analyses like correlation and classification. Our results indicate that the proposed network approach can improve accuracy of classifying subjects into two groups (control and TLE) from 78% for non-network classifiers to 93% using the proposed approach. We also obtain network "peakiness" values using statistical measures like entropy and complexity-this appears to be a good characterizer of the disease and may have utility in surgical planning.

Histology classification is not a predictor of clinical outcomes in advanced non-small cell lung cancer (NSCLC) treated with vinorelbine or gemcitabine combinations

BACKGROUND

Until recently, histology has not been clearly or consistently described in the literature as a prognostic or predictive variable in advanced NSCLC studies. We have categorised patients treated with vinorelbine and gemcitabine based first line chemotherapy regimes for advanced NSCLC as either squamous or non-squamous, and also as either adenocarcinoma and non-adenocarcinoma, and compared outcome.

MATERIAL AND METHODS

420 patients treated with platinum/gemcitabine, platinum/vinorelbine or single agent gemcitabine or vinorelbine as first line chemotherapy for advanced NSCLC were identified. The influence of pathology on progression free survival (PFS) and overall survival (OS) has been investigated by means of a Cox regression analysis. Hazard ratios with 95% CIs have been given for each pathological type after adjusting for the effects of age, gender, stage (III vs. IV), PS (0/1 vs. 2/3) and treatment type (platinum doublet vs. single agent).

RESULTS

Neither univariate nor multivariate analysis suggested that there was a significant difference in the response rates for adenocarcinoma vs. non-adenocarcinoma or between squamous and non-squamous pathology. There was no difference in PFS between adenocarcinoma and non-adenocarcinoma pathologies until 8 months (p = 0.98), and there was a statistically significant advantage in PFS for squamous vs. non-squamous pathologies (p = 0.04). Using multivariate Cox regression analysis to adjust for the effects of age, gender, stage, PS, and treatment type, the pathology subtype was not significant. There was no difference in OS in any group.

CONCLUSIONS

These results suggest that histology may not be considered as a predictor of clinical outcome using these drugs.

Cyber disorders: the mental health concern for the new millennium

Anecdotal evidence has suggested that mental health practitioners' report increased caseloads of clients whose primary complaint involves the Internet. However, little is known about the incidence, associated behaviors, attitudes of practitioners, and interventions involved related to this relatively new phenomenon. Therefore, this study surveyed therapists who have treated clients suffering from cyber-related problems to gather such outcome information. Respondents reported an average caseload of nine clients who they classified as Internet-addicted, with a range between 2 and 50 clients treated within the past year. Five general subtypes of Internet addiction were categorized based on the most problematic types of online applications, and they include addictions to Cybersex, Cyber-relationships, online stock trading or gambling, information surfing, and computer games. Treatment strategies included cognitive-behavioral approaches, sexual offender therapy, marital and family therapy, social skills training, and pharmacological interventions. Based on their client encounters, efforts to initiate support groups and recovery programs specializing in the treatment of Internet addiction were being considered. Finally, based upon the findings, this article examines the impact of cyberdisorders on future research, treatment, and public policy issues for the new millennium.

Effective consulting and history-taking skills for prescribing practice

There is increasing emphasis on the need for nurses to develop history-taking and consultation skills for prescribing and advanced clinical practice roles. This article discusses both theoretical and practical aspects that could facilitate the development of these skills using safe and structured approaches. It explores some of the origins of consultation theory, and the challenge that nurses face in integrating new consultation skills with existing nursing assessment practice. This article outlines practical generic approaches to history taking in the clinical consultation and considers the main areas that need to be covered to support safe prescribing decisions.

An alternative method for noise analysis using pixel variance as part of quality control procedures on digital mammography systems

According to the European Guidelines for quality assured breast cancer screening and diagnosis, noise analysis is one of the measurements that needs to be performed as part of quality control procedures on digital mammography systems. However, the method recommended in the European Guidelines does not discriminate sufficiently between systems with and without additional noise besides quantum noise. This paper attempts to give an alternative and relatively simple method for noise analysis which can divide noise into electronic noise, structured noise and quantum noise. Quantum noise needs to be the dominant noise source in clinical images for optimal performance of a digital mammography system, and therefore the amount of electronic and structured noise should be minimal. For several digital mammography systems, the noise was separated into components based on the measured pixel value, standard deviation (SD) of the image and the detector entrance dose. The results showed that differences between systems exist. Our findings confirm that the proposed method is able to discriminate systems based on their noise performance and is able to detect possible quality problems. Therefore, we suggest to replace the current method for noise analysis as described in the European Guidelines by the alternative method described in this paper.

Assessing a drop box programme: a spatial analysis of discarded needles

BACKGROUND

Distributing sterile injection equipment to injection drug users is one of few proven ways of lowering the transmission rate of blood borne viruses. Distribution of equipment has also been linked to increased needle discarding, which is a public health risk for both injectors and their host communities. Drop boxes (anonymous and public-access sharps containers) are a promising and increasingly popular means of reducing unsafe disposal, yet there is little empirical research to support or guide their implementation.

METHODS

Using a dataset containing the locations of 7274 discarded needles and syringes collected monthly in the non-park open spaces of a 2.5km(2) neighbourhood of Montréal, Canada for a period of five years, we compared levels of discards before and after the installation of 12 drop boxes. We used quasi-Poisson regression to test the effects of drop boxes on monthly counts of collected discards for areas within a walking distance of 25, 50, 100 and 200m of a drop box. We adjusted for known time-dependent covariates linearly and unknown time-dependent covariates using a smoothing function.

RESULTS

We found strong evidence of reduced discarding following the installation of drop boxes; drop boxes were associated with reductions of up to 98% (95% CI: 72-100%) and significant reductions for areas up to 200m from a drop box. Reductions were inversely proportional to walking distance from drop boxes. No measure of weather or use of needle exchange programmes (NEPs) had a consistent relationship with discard counts.

CONCLUSION

Our research suggests that IDUs changed their needle-disposal behaviour in response to increased safe disposal options. In addition to being relatively low-threshold, economical and rapid, drop boxes appear to be a highly effective intervention to reduce discarded needles.

Galantamine-induced improvements in cognitive function are not related to alterations in alpha(4)beta (2) nicotinic receptors in early Alzheimer's disease as measured in vivo by 2-[18F]fluoro-A-85380 PET

INTRODUCTION

The nicotinic acetylcholine receptor (nAChR) system plays a regulatory role in a number of cognitive processes. Cholinesterase inhibitors (i.e., galantamine) that potentiate cholinergic neurotransmission improve cognitive function in Alzheimer's disease (AD); however, the relationship between these effects and associated changes in nAChRs are yet to be established in vivo.

MATERIALS AND METHODS

2-[18F]Fluoro-A-85380 (2-FA) binds to nAChRs and with positron emission tomography (PET) imaging provides a composite measure of receptor density and ligand affinity. This study aimed to: (1) quantify nAChRs in vivo in 15 drug-naïve patients with mild AD before and after chronic treatment with galantamine, using 2-FA and PET, and (2) examine the relationship between treatment-induced changes in nAChRs and improvements in cognitive function. Participants were nonsmokers and underwent extensive cognitive testing and a PET scan after injection of approximately 200 MBq of 2-FA on two occasions (before and after 12 weeks, galantamine treatment). A 3-day washout period preceded the second scan. Brain regional 2-FA binding was assessed through a simplified estimation of distribution volume (DV(S)).

RESULTS

Performance on global measures of cognition significantly improved following galantamine treatment (p < 0.05). This improvement extended to specific cognitive measures of language and verbal learning. No significant differences in nAChR DV(S) before and after galantamine treatment were found. The treatment-induced improvement in cognition was not correlated with regional or global nAChR DV(S), suggesting that changes in nAChRs may not be responsible for the improvements in cognition following galantamine in patients with mild AD.

Vitamin D insufficiency and treatment with oral vitamin D3 in northern-dwelling patients with chronic kidney disease

BACKGROUND

Vitamin D insufficiency is common in people living at northern latitudes and those with chronic kidney disease (CKD). We studied persons with both of these risk factors to determine the prevalence of vitamin D insufficiency and whether serum 25-hydroxyvitamin D (25(OH)D) levels were affected by oral vitamin D3 supplementation.

METHODS

This was a prospective controlled trial of 128 patients with stage 3-5 non-dialysis dependent CKD. Patients were assigned to the intervention (oral vitamin D3 1,000 IU/day) in a 1:1 ratio at the discretion of the attending dietitian. Serum biochemical markers were measured at baseline (May-July) and after 3 months of follow-up. There were 63 control and 65 intervention subjects.

RESULTS

Mean 25(OH)D levels increased significantly higher in the treatment group (mean increase from baseline: 10.3+/-10.4 ng/mL vs. 0.8+/-6.8 ng/mL, p<0.0001). This difference remained significant after adjustment for differing baseline characteristics between groups (p<0.0001). Treatment with oral vitamin D3 reduced vitamin D insufficiency by 37%, as compared with a 2% increase in prevalence among the control group (p<0.0001). Considering the entire study population, 93% of patients had levels less than <30 ng/mL at least once during the study.

CONCLUSION

Vitamin D insufficiency is highly prevalent in northern-dwelling patients with stage 3-5 CKD, and is moderated by oral supplementation with 1,000 IU of vitamin D3 daily.

Non-invasive ventilation in chronic obstructive pulmonary disease: management of acute type 2 respiratory failure

Non-invasive ventilation (NIV) in the management of acute type 2 respiratory failure in patients with chronic obstructive pulmonary disease (COPD) represents one of the major technical advances in respiratory care over the last decade. This document updates the 2002 British Thoracic Society guidance and provides a specific focus on the use of NIV in COPD patients with acute type 2 respiratory failure. While there are a variety of ventilator units available most centres now use bi-level positive airways pressure units and this guideline refers specifically to this form of ventilatory support although many of the principles encompassed are applicable to other forms of NIV. The guideline has been produced for the clinician caring for COPD patients in the emergency and ward areas of acute hospitals.

An automated communication system in a contact registry for persons with rare diseases: scalable tools for identifying and recruiting clinical research participants

OBJECTIVES

Strategies for study recruitment are useful in clinical research network settings. We describe a registry of individuals who have self-identified with one of a multiplicity of rare diseases, and who express a willingness to be contacted regarding possible enrollment in clinical research studies. We evaluate this registry and supporting tools in terms of registry enrollment and impact on participation rates in advertised clinical research studies.

METHODS

A web-based automated system generates periodic and customized communications to notify registrants of relevant studies in the NIH Rare Diseases Clinical Research Network (RDCRN). The majority of these communications are sent by email. We compare the characteristics of those enrolled in the registry to the characteristics of participants enrolled in sampled RDCRN studies in order to estimate the impact of the registry on study participation in the network.

RESULTS

The registry currently contains over 4000 registrants, representing 40 rare diseases. Estimates of study participation range from 6-27% for all enrollees. Study participation rates for some disease areas are over 40% when considering only contact registry enrollees who live within 100 mi of a clinical research study site.

CONCLUSIONS

Automated notifications can facilitate consistent, customized, and timely communication of relevant protocol information to potential research subjects. Our registry and supporting communication tools demonstrate a significant positive impact on study participation rates in our network. The use of the internet and automated notifications make the system scalable to support many protocols and registrants.