Osama Mahmoud

A feature selection method for classification within functional genomics experiments based on the proportional overlapping score.

BackgroundMicroarray technology, as well as other functional genomics experiments, allow simultaneous measurements of thousands of genes within each sample. Both the prediction accuracy and interpretability of a classifier could be enhanced by performing the classification based only on selected discriminative genes. We propose a statistical method for selecting genes based on overlapping analysis of expression data across classes. This method results in a novel measure, called proportional overlapping score (POS), of a feature’s relevance to a classification task.ResultsWe apply POS, along‐with four widely used gene selection methods, to several benchmark gene expression datasets. The experimental results of classification error rates computed using the Random Forest, k Nearest Neighbor and Support Vector Machine classifiers show that POS achieves a better performance.ConclusionsA novel gene selection method, POS, is proposed. POS analyzes the expressions overlap across classes taking into account the proportions of overlapping samples. It robustly defines a mask for each gene that allows it to minimize the effect of expression outliers. The constructed masks along‐with a novel gene score are exploited to produce the selected subset of genes.

Ensemble of a subset of kNN classifiers

Combining multiple classifiers, known as ensemble methods, can give substantial improvement in prediction performance of learning algorithms especially in the presence of non-informative features in the data sets. We propose an ensemble of subset of kNN classifiers, ESkNN, for classification task in two steps. Firstly, we choose classifiers based upon their individual performance using the out-of-sample accuracy. The selected classifiers are then combined sequentially starting from the best model and assessed for collective performance on a validation data set. We use bench mark data sets with their original and some added non-informative features for the evaluation of our method. The results are compared with usual kNN, bagged kNN, random kNN, multiple feature subset method, random forest and support vector machines. Our experimental comparisons on benchmark classification problems and simulated data sets reveal that the proposed ensemble gives better classification performance than the usual kNN and its ensembles, and performs comparable to random forest and support vector machines.

Centre-level variation of treatment and outcome in 5-year-old children with non-syndromic unilateral cleft lip and palate: the Cleft Care UK study. Part 1: Methodology and results for dento-facial outcomes

Structured Abstract Objectives Outline methods used to describe centre‐level variation in treatment and outcome in children in the Cleft Care UK (CCUK) study. Report centre‐level variation in dento‐facial outcomes. Setting and Sample Population Two hundred and sixty‐eight five‐year‐old British children with non‐syndromic unilateral cleft lip and palate (UCLP). Materials and Methods Between January 2011 and December 2012, data were collected on a comprehensive range of outcomes. Child facial appearance and symmetry were assessed using photographic pictures. Dental arch relationships were assessed from standardized dental study models. Hierarchical statistical models were used to predict overall means and the variance partition coefficient (VPC)—a measure of amount of variation in treatment or outcome explained by the centre. Results Data on dento‐alveolar arch relationships and facial appearance were available on 197 and 252 children, respectively. The median age of the children was 5.5 years, and 68% were boys. Variation was described across 13 centres. There was no evidence of centre‐level variation in good or poor dento‐alveolar arch relationships with a VPC of 4% and 3%, respectively. Similarly, there was no evidence of centre‐level variation in good or poor facial appearance with a VPC of 2% and 5%, respectively. Conclusions There was no evidence of centre‐level variation for dento‐facial outcomes although this study only had the power to detect large variation between sites.

Centre-level variation in speech outcome and interventions, and factors associated with poor speech outcomes in 5-year-old children with non-syndromic unilateral cleft lip and palate: The Cleft Care UK study. Part 4

OBJECTIVESnTo investigate centre-level variation in speech intervention and outcome and factors associated with a speech disorder in children in Cleft Care UK (CCUK).nnnSETTING AND SAMPLE POPULATIONnTwo hundred and sixty-eight 5-year-old British children with non-syndromic unilateral cleft lip and palate recruited to CCUK.nnnMATERIALS AND METHODSnCentre-based therapists undertook audio-video recordings. Perceptual analysis was undertaken using the CAPS-A tool. Speech outcomes were based on structural and articulation scores, and intelligibility/distinctiveness. Between-centre variation in treatment and outcomes were examined using multilevel models. These models were extended to estimate the association between a range of factors (hearing loss, speech intervention, fistula, secondary speech surgery for velopharyngeal insufficiency, socio-economic status, gender, and parental happiness with speech) and speech outcomes.nnnRESULTSnThere was centre-level variation in secondary speech surgery, speech intervention, structure and intelligibility outcomes. Children with a history of speech intervention had a lower odds of poor intelligibility/distinctiveness, 0.1 (95% CI: 0.0-0.4). Parental concern was associated with a higher odds of poor intelligibility/distinctiveness, 13.2 (95% CI: 4.9-35.1). Poor speech outcomes were associated with a fistula, secondary speech surgery and history of hearing loss.nnnCONCLUSIONSnWithin the centralized service there is centre-level variation in secondary speech surgery, intervention and speech outcomes. These findings support the importance of early management of fistulae, effective management of velopharyngeal insufficiency and hearing impairment, and most importantly speech intervention in the preschool years. Parental concern about speech is a good indicator of speech status.

Centre-level variation in behaviour and the predictors of behaviour in 5-year-old children with non-syndromic unilateral cleft lip: The Cleft Care UK study. Part 5

Structured Abstract Objectives The aims of this study were to describe child behavioural and psychosocial outcomes associated with appearance and speech in the Cleft Care UK (CCUK) study. We also wanted to explore centre‐level variation in child outcomes and investigate individual predictors of such outcomes. Setting and sample population Two hundred and sixty‐eight five‐year‐old children with non‐syndromic unilateral cleft lip and palate (UCLP) recruited to CCUK. Materials and methods Parents completed the Strengths and Difficulties questionnaire (SDQ) and reported their own perceptions of the childs self‐confidence. Child facial appearance and symmetry were assessed using photographs, and intelligibility of speech was derived from audio‐visual speech recordings. Centre‐level variation in behavioural and psychosocial outcomes was examined using hierarchical models, and associations with clinical outcomes were examined using logit regression models. Results Children with UCLP had a higher hyperactive difficulty score than the general population. For boys, the average score was 4.5 vs 4.1 (P=.03), and for girls, the average score was 3.8 vs 3.1 (P=.008). There was no evidence of centre‐level variation for behaviour or parental perceptions of the childs self‐confidence. There is no evidence of associations between self‐confidence and SDQ scores and either facial appearance or behaviour. Conclusions Children born with UCLP have higher levels of behaviour problems than the general population.

An Ensemble of Optimal Trees for Class Membership Probability Estimation

Machine learning methods can be used for estimating the class membership probability of an observation. We propose an ensemble of optimal trees in terms of their predictive performance. This ensemble is formed by selecting the best trees from a large initial set of trees grown by random forest. A proportion of trees is selected on the basis of their individual predictive performance on out-of-bag observations. The selected trees are further assessed for their collective performance on an independent training data set. This is done by adding the trees one by one starting from the highest predictive tree. A tree is selected for the final ensemble if it increases the predictive performance of the previously combined trees. The proposed method is compared with probability estimation tree, random forest and node harvest on a number of bench mark problems using Brier score as a performance measure. In addition to reducing the number of trees in the ensemble, our method gives better results in most of the cases. The results are supported by a simulation study.

Centre-level variation in treatment and outcomes and predictors of outcomes in 5-year-old children with non-syndromic unilateral cleft lip treated within a centralized service: The Cleft Care UK study. Part 6: Summary and implications

OBJECTIVESnTo summarize and discuss centre-level variation across a range of treatment and outcome measures and examine individual and ecological determinants of outcome in children in Cleft Care UK (CCUK).nnnSETTING AND SAMPLE POPULATIONnTwo hundred and sixty-eight 5-year-old British children with non-syndromic unilateral cleft lip and palate (UCLP) recruited to CCUK and treated within a centralized service.nnnMATERIALS AND METHODSnChildren had a range of treatment and outcome measures collected at a 5-year audit clinic. These outcomes included dento-alveolar arch relationships from study models, measures of facial appearance from cropped photographs, hearing loss from audiological assessment, speech from speech recordings, self-confidence and strengths and difficulties from parental self-report. Data were collected on educational attainment at age 7 using record linkage. Centre variation was examined using hierarchical regression and associations between variables were examined using logistic or poisson regression.nnnRESULTSnThere was centre-level variation for some treatments (early grommet placement, fitting of hearing aids, fluoride treatment, secondary speech surgery and treatment for cleft speech characteristics) and for some outcomes (intelligibility of speech). Hearing loss was associated with a higher risk of poor speech while speech therapy was associated with a lower risk of poor speech. Children had high levels of caries but levels of preventative treatment (fluoride varnish and tablets) were low.nnnCONCLUSIONSnFurther improvements to and monitoring of the current centralized model of care are required to ensure the best outcomes for all children with cleft lip and palate.

Hypercholesterolaemia screening in Type 1 diabetes: a difference of opinion

To assess cholesterol screening of children with Type 1 diabetes by diabetes professionals using a survey of current practice, given that National Institute of Health and Care Excellence guidelines on childhood Type 1 diabetes do not recommend cholesterol screening, yet the National Paediatric Diabetes Audit has an annual cholesterol measure (> 12 years) as a key outcome indicator.

Treatment adherence and BMI reduction are key predictors of HbA1c one year after diagnosis of childhood Type 2 Diabetes in UK

Type 2 Diabetes (T2DM) is increasing in childhood especially among females and South‐Asians. Our objective was to report outcomes from a national cohort of children and adolescents with T2DM 1 year following diagnosis.

Age at puberty and risk of asthma: A Mendelian randomisation study

Background Observational studies on pubertal timing and asthma, mainly performed in females, have provided conflicting results about a possible association of early puberty with higher risk of adult asthma, possibly due to residual confounding. To overcome issues of confounding, we used Mendelian randomisation (MR), i.e., genetic variants were used as instrumental variables to estimate causal effects of early puberty on post-pubertal asthma in both females and males. Methods and findings MR analyses were performed in UK Biobank on 243,316 women using 254 genetic variants for age at menarche, and on 192,067 men using 46 variants for age at voice breaking. Age at menarche, recorded in years, was categorised as early (<12), normal (12–14), or late (>14); age at voice breaking was recorded and analysed as early (younger than average), normal (about average age), or late (older than average). In females, we found evidence for a causal effect of pubertal timing on asthma, with an 8% increase in asthma risk for early menarche (odds ratio [OR] 1.08; 95% CI 1.04 to 1.12; p = 8.7 × 10−5) and an 8% decrease for late menarche (OR 0.92; 95% CI 0.89 to 0.97; p = 3.4 × 10−4), suggesting a continuous protective effect of increasing age at puberty. In males, we found very similar estimates of causal effects, although with wider confidence intervals (early voice breaking: OR 1.07; 95% CI 1.00 to 1.16; p = 0.06; late voice breaking: OR 0.93; 95% CI 0.87 to 0.99; p = 0.03). We detected only modest pleiotropy, and our findings showed robustness when different methods to account for pleiotropy were applied. BMI may either introduce pleiotropy or lie on the causal pathway; secondary analyses excluding variants associated with BMI yielded similar results to those of the main analyses. Our study relies on self-reported exposures and outcomes, which may have particularly affected the power of the analyses on age at voice breaking. Conclusions This large MR study provides evidence for a causal detrimental effect of early puberty on asthma, and does not support previous observational findings of a U-shaped relationship between pubertal timing and asthma. Common biological or psychological mechanisms associated with early puberty might explain the similarity of our results in females and males, but further research is needed to investigate this. Taken together with evidence for other detrimental effects of early puberty on health, our study emphasises the need to further investigate and address the causes of the secular shift towards earlier puberty observed worldwide.