A tree-based decision model to support prediction of the severity of asthma exacerbations in children

This paper describes the development of a tree-based decision model to predict the severity of pediatric asthma exacerbations in the emergency department (ED) at 2 h following triage. The model was constructed from retrospective patient data abstracted from the ED charts. The original data was preprocessed to eliminate questionable patient records and to normalize values of age-dependent clinical attributes. The model uses attributes routinely collected in the ED and provides predictions even for incomplete observations. Its performance was verified on independent validating data (split-sample validation) where it demonstrated AUC (area under ROC curve) of 0.83, sensitivity of 84%, specificity of 71% and the Brier score of 0.18. The model is intended to supplement an asthma clinical practice guideline, however, it can be also used as a stand-alone decision tool.

Clinical Decision Support System Sonares

A decision support system SonaRes destined to guide and help the ultrasound operators is proposed and compared with the existing ones. The system is based on rules and images and can be used as a second opinion in the process of ultrasound examination.

Derivation of Context-free Stochastic L-Grammar Rules for Promoter Sequence Modeling Using Support Vector Machine

Formal grammars can used for describing complex repeatable structures such as DNA sequences. In this paper, we describe the structural composition of DNA sequences using a context-free stochastic L-grammar. L-grammars are a special class of parallel grammars that can model the growth of living organisms, e.g. plant development, and model the morphology of a variety of organisms. We believe that parallel grammars also can be used for modeling genetic mechanisms and sequences such as promoters. Promoters are short regulatory DNA sequences located upstream of a gene. Detection of promoters in DNA sequences is important for successful gene prediction. Promoters can be recognized by certain patterns that are conserved within a species, but there are many exceptions which makes the promoter recognition a complex problem. We replace the problem of promoter recognition by induction of context-free stochastic L-grammar rules, which are later used for the structural analysis of promoter sequences. L-grammar rules are derived automatically from the drosophila and vertebrate promoter datasets using a genetic programming technique and their fitness is evaluated using a Support Vector Machine (SVM) classifier. The artificial promoter sequences generated using the derived L- grammar rules are analyzed and compared with natural promoter sequences.

Intraoperative prediction of refractive error using a prototype surgical aberrometer

Purpose: To ascertain the agreement level between intra-operative refraction using a prototype surgical Hartmann-Shack aberrometer and subjective refraction a month later. Methods: Fifty-four consecutive patients had their pseudophakic refractive measured with the aberrometer intra-operatively at the end of their cataract surgery. A masked optometrist performed subjective refraction 4 weeks later. The two sets of data were then analysed for correlation. Results: The mean spherical equivalent was −0.14 ± 0.37 D (Range: −1.41 to +1.72 D) with the prototype aberrometer and −0.34 ± 0.32 (−1.64 to +1.88 D) with subjective refraction. The measurements positively correlated to a very high degree (r =+0.81, p < 0.01). In 84.3% of cases the two measurements were within 0.50D of each other. Conclusion: The aberrometer can verify the aimed refractive status of the eye intraoperatively to avoid a refractive surprise. The aberrometer is a useful tool for real time assessment of the ocular refractive status.

Towards the prediction of mutations in genomic sequences

Bio-systems are inherently complex information processing systems. Furthermore, physiological complexities of biological systems limit the formation of a hypothesis in terms of behavior and the ability to test hypothesis. More importantly the identification and classification of mutation in patients are centric topics in today's cancer research. Next generation sequencing (NGS) technologies can provide genome-wide coverage at a single nucleotide resolution and at reasonable speed and cost. The unprecedented molecular characterization provided by NGS offers the potential for an individualized approach to treatment. These advances in cancer genomics have enabled scientists to interrogate cancer-specific genomic variants and compare them with the normal variants in the same patient. Analysis of this data provides a catalog of somatic variants, present in tumor genome but not in the normal tissue DNA. In this dissertation, we present a new computational framework to the problem of predicting the number of mutations on a chromosome for a certain patient, which is a fundamental problem in clinical and research fields. We begin this dissertation with the development of a framework system that is capable of utilizing published data from a longitudinal study of patients with acute myeloid leukemia (AML), who's DNA from both normal as well as malignant tissues was subjected to NGS analysis at various points in time. By processing the sequencing data at the time of cancer diagnosis using the components of our framework, we tested it by predicting the genomic regions to be mutated at the time of relapse and, later, by comparing our results with the actual regions that showed mutations (discovered at relapse time). We demonstrate that this coupling of the algorithm pipeline can drastically improve the predictive abilities of searching a reliable molecular signature. Arguably, the most important result of our research is its superior performance to other methods like Radial Basis Function Network, Sequential Minimal Optimization, and Gaussian Process. In the final part of this dissertation, we present a detailed significance, stability and statistical analysis of our model. A performance comparison of the results are presented. This work clearly lays a good foundation for future research for other types of cancer.^

Occupant location prediction in smart buildings using association rule mining

Heating, ventilation, air conditioning (HVAC) systems are significant consumers of energy, however building management systems do not typically operate them in accordance with occupant movements. Due to the delayed response of HVAC systems, prediction of occupant locations is necessary to maximize energy efficiency. We present an approach to occupant location prediction based on association rule mining, allowing prediction based on historical occupant locations. Association rule mining is a machine learning technique designed to find any correlations which exist in a given dataset. Occupant location datasets have a number of properties which differentiate them from the market basket datasets that association rule mining was originally designed for. This thesis adapts the approach to suit such datasets, focusing the rule mining process on patterns which are useful for location prediction. This approach, named OccApriori, allows for the prediction of occupants’ next locations as well as their locations further in the future, and can take into account any available data, for example the day of the week, the recent movements of the occupant, and timetable data. By integrating an existing extension of association rule mining into the approach, it is able to make predictions based on general classes of locations as well as specific locations.

The Comparative Effectiveness of EEG Biomarkers in Antidepressant Response and Illness Prediction in Major Depressive Disorder

Background: There is growing evidence that individual EEG differences may aid in classifying patients with major depressive disorder (MDD) and also help predict clinical response to antidepressant treatment. This study aims to compare the effectiveness of EEG frequency band power, alpha asymmetry and prefrontal theta cordance towards escitalopram response prediction and MDD diagnosis, in a multi-site initiative. Methods: Resting EEG (eyes open and closed) was recorded from 64 electrodes in 44 depressed patients and 20 healthy controls at baseline, 2 weeks post-treatment and 8 weeks post-treatment. Clinical response was measured as change from baseline MADRS of 50% or more. EEG measures were analyzed (1) at baseline (2) at 2 weeks post-treatment and (3) as an ‘‘early change” variable defined as change in EEG from baseline to 2 weeks post-treatment. Results: At baseline, responders exhibited greater absolute alpha power in the left hemisphere versus the right while non-responders showed the opposite. Responders further exhibited a cortical asymmetry of greater right relative to left activity in parietal areas. Groups also differed in baseline relative delta power with responders showing greater power in the right hemisphere versus the left while non-responders showed the opposite. At 2 weeks post-treatment, responders exhibited greater absolute beta power in the left hemisphere relative to right and the opposite was noted for non-responders. The opposite pattern was noted for absolute and relative delta power at 2 weeks post-treatment. Responders exhibited early reduction in relative alpha power and early increments in relative theta power. Non-responders showed a significant early increase in prefrontal theta cordance. Absolute delta power helped distinguish MDD patients from healthy controls. Conclusions: Hemispheric asymmetries in the alpha and delta bands at pre-treatment baseline and at 2 weeks post-treatment have moderate to moderately strong predictive utility towards antidepressant treatment response. These findings have significant potential for improving clinical practice in psychiatry by eventually guiding clinical choice of treatments. This would greatly benefit patients awaiting relief from depressive symptoms as treatment optimization would help overcome problems associated with delayed recovery. Our results also indicate that resting EEG activity may have clinical utility in predicting MDD diagnosis.

Troponin I and cardiovascular risk prediction in the general population: the BiomarCaRE consortium

AIMS: Our aims were to evaluate the distribution of troponin I concentrations in population cohorts across Europe, to characterize the association with cardiovascular outcomes, to determine the predictive value beyond the variables used in the ESC SCORE, to test a potentially clinically relevant cut-off value, and to evaluate the improved eligibility for statin therapy based on elevated troponin I concentrations retrospectively.

METHODS AND RESULTS: Based on the Biomarkers for Cardiovascular Risk Assessment in Europe (BiomarCaRE) project, we analysed individual level data from 10 prospective population-based studies including 74 738 participants. We investigated the value of adding troponin I levels to conventional risk factors for prediction of cardiovascular disease by calculating measures of discrimination (C-index) and net reclassification improvement (NRI). We further tested the clinical implication of statin therapy based on troponin concentration in 12 956 individuals free of cardiovascular disease in the JUPITER study. Troponin I remained an independent predictor with a hazard ratio of 1.37 for cardiovascular mortality, 1.23 for cardiovascular disease, and 1.24 for total mortality. The addition of troponin I information to a prognostic model for cardiovascular death constructed of ESC SCORE variables increased the C-index discrimination measure by 0.007 and yielded an NRI of 0.048, whereas the addition to prognostic models for cardiovascular disease and total mortality led to lesser C-index discrimination and NRI increment. In individuals above 6 ng/L of troponin I, a concentration near the upper quintile in BiomarCaRE (5.9 ng/L) and JUPITER (5.8 ng/L), rosuvastatin therapy resulted in higher absolute risk reduction compared with individuals <6 ng/L of troponin I, whereas the relative risk reduction was similar.

CONCLUSION: In individuals free of cardiovascular disease, the addition of troponin I to variables of established risk score improves prediction of cardiovascular death and cardiovascular disease.

Common polygenic variation enhances risk prediction for Alzheimer's disease.

The identification of subjects at high risk for Alzheimer’s disease is important for prognosis and early intervention. We investigated the polygenic architecture of Alzheimer’s disease and the accuracy of Alzheimer’s disease prediction models, including and excluding the polygenic component in the model. This study used genotype data from the powerful dataset comprising 17 008 cases and 37 154 controls obtained from the International Genomics of Alzheimer’s Project (IGAP). Polygenic score analysis tested whether the alleles identified to associate with disease in one sample set were significantly enriched in the cases relative to the controls in an independent sample. The disease prediction accuracy was investigated in a subset of the IGAP data, a sample of 3049 cases and 1554 controls (for whom APOE genotype data were available) by means of sensitivity, specificity, area under the receiver operating characteristic curve (AUC) and positive and negative predictive values. We observed significant evidence for a polygenic component enriched in Alzheimer’s disease (P = 4.9 × 10−26). This enrichment remained significant after APOE and other genome-wide associated regions were excluded (P = 3.4 × 10−19). The best prediction accuracy AUC = 78.2% (95% confidence interval 77–80%) was achieved by a logistic regression model with APOE, the polygenic score, sex and age as predictors. In conclusion, Alzheimer’s disease has a significant polygenic component, which has predictive utility for Alzheimer’s disease risk and could be a valuable research tool complementing experimental designs, including preventative clinical trials, stem cell selection and high/low risk clinical studies. In modelling a range of sample disease prevalences, we found that polygenic scores almost doubles case prediction from chance with increased prediction at polygenic extremes.

Risk Prediction In Cardiovascular Epidemiology: Considerations for Modeling Composite Endpoints

Thesis (Master's)--University of Washington, 2016-08

Renal Drug Transporters – Clinical Significance and Role in Drug-Drug Interactions

Thesis (Ph.D.)--University of Washington, 2016-07

A long-term energy efficiency prediction model for the Brazilian automotive industry

According to law number 12.715/2012, Brazilian government instituted guidelines for a program named Inovar-Auto. In this context, energy efficiency is a survival requirement for Brazilian automotive industry from September 2016. As proposed by law, energy efficiency is not going to be calculated by models only. It is going to be calculated by the whole universe of new vehicles registered. In this scenario, the composition of vehicles sold in market will be a key factor on profits of each automaker. Energy efficiency and its consequences should be taken into consideration in all of its aspects. In this scenario, emerges the following question: which is the efficiency curve of one automaker for long term, allowing them to adequate to rules, keep balancing on investment in technologies, increasing energy efficiency without affecting competitiveness of product lineup? Among several variables to be considered, one can highlight the analysis of manufacturing costs, customer value perception and market share, which characterizes this problem as a multi-criteria decision-making. To tackle the energy efficiency problem required by legislation, this paper proposes a framework of multi-criteria decision-making. The proposed framework combines Delphi group and Analytic Hierarchy Process to identify suitable alternatives for automakers to incorporate in main Brazilian vehicle segments. A forecast model based on artificial neural networks was used to estimate vehicle sales demand to validate expected results. This approach is demonstrated with a real case study using public vehicles sales data of Brazilian automakers and public energy efficiency data.

Performance of protective factors assessment in risk prediction for adults: systematic review and meta-analysis

Protective factors are neglected in risk assessment in adult psychiatric and criminal justice populations. This review investigated the predictive efficacy of selected tools that assess protective factors. Five databases were searched using comprehensive terms for records up to June 2014, resulting in 17 studies (n = 2,198). Results were combined in a multilevel meta-analysis using the R (R Core Team, R: A Language and Environment for Statistical Computing, Vienna, Austria: R Foundation for Statistical Computing, 2015) metafor package (Viechtbauer, Journal of Statistical Software, 2010, 36, 1). Prediction of outcomes was poor relative to a reference category of violent offending, with the exception of prediction of discharge from secure units. There were no significant differences between the predictive efficacy of risk scales, protective scales, and summary judgments. Protective factor assessment may be clinically useful, but more development is required. Claims that use of these tools is therapeutically beneficial require testing.

An investigation of the clinical utility of preoperative cardiopulmonary exercise testing in patients undergoing major pancreatic surgery

Pancreaticoduodenectomy with or without adjuvant chemotherapy remains the only modality of possible cure in patients with cancer involving the head of the pancreas and the periampullary region. While mortality rates after pancreaticoduodenectomy have improved considerably over the course of the last century, morbidity remains high. Patient selection is of paramount importance in ensuring that major surgery is offered to individuals who will most benefit from a pancreaticoduodenectomy. Moreover, identifying preoperative risk factors provides potential targets for prehabilitation and optimisation of the patient's physiology before undertaking surgery. In addition to this, early identification of patients who are likely to develop postoperative complications allows for better allocation of critical care resources and more aggressive management high risk patients. Cardiopulmonary exercise testing is becoming an increasingly popular tool in the preoperative risk assessment of the surgical patient. However, very little work has been done to investigate the role of cardiopulmonary exercise testing in predicting complications after pancreaticoduodenectomy. The impact of jaundice, systemic inflammation and other preoperative clinicopathological characteristics on cardiopulmonary exercise physiology has not been studied in detail before in this cohort of patients. The overall aim of the thesis was to examine the relationships between preoperative clinico-pathological characteristics including cardiopulmonary exercise physiology, obstructive jaundice, body composition and systemic inflammation and complications and the post-surgical systemic inflammatory response in patients undergoing pancreaticoduodenectomy. Chapter 1 reviews the existing literature on preoperative cardiopulmonary exercise testing, the impact of obstructive jaundice, perioperative systemic inflammation and the importance of body composition in determining outcomes in patients undergoing major surgery with particular reference to pancreatic surgery. Chapter 2 reports on the role of cardiopulmonary exercise testing in predicting postoperative complications after pancreaticoduodenectomy. The results demonstrate that patients with V˙O2AT less than 10 ml/kg/min are more likely to develop a postoperative pancreatic fistula, stay longer in hospital and less likely to receive adjuvant therapy. These results emphasise the importance of aerobic fitness to recover from the operative stress of major surgery without significant morbidity. Cardiopulmonary exercise testing may prove useful in selecting patients for intensive prehabilitation programmes as well as for other optimisation measures to prepare them for major surgery. Chapter 3 evaluates the relationship between cardiopulmonary exercise physiology and other clinicopathological characteristics of the patient. A detailed analysis of cardiopulmonary exercise test parameters in jaundiced versus non-jaundiced patients demonstrates that obstructive jaundice does not impair cardiopulmonary exercise physiology. This further supports emerging evidence in contemporary literature that jaundiced patients can proceed directly to surgery without preoperative biliary drainage. The results of this study also show an interesting inverse relationship between body mass index and anaerobic threshold which is analysed in more detail in Chapter 4. Chapter 4 examines the relationship between preoperative cardiopulmonary exercise physiology and body composition in depth. All parameters measured at cardiopulmonary exercise test are compared against body composition and body mass index. The results of this chapter report that the current method of reporting V˙O2, both at peak exercise and anaerobic threshold, is biased against obese subjects and advises caution in the interpretation of cardiopulmonary exercise test results in patients with a high BMI. This is particularly important as current evidence in literature suggests that postoperative outcomes in obese subjects are comparable to non-obese subjects while cardiopulmonary exercise test results are also abnormally low in this very same cohort of patients. Chapter 5 analyses the relationship between preoperative clinico-pathological characteristics including systemic inflammation and the magnitude of the postoperative systemic inflammatory response. Obstructive jaundice appears to have an immunosuppressive effect while elevated preoperative CRP and hypoalbuminemia appear to have opposite effects with hypoalbuminemia resulting in a lower response while elevated CRP in the absence of hypoalbuminemia resulted in a greater postoperative systemic inflammatory response. Chapter 6 evaluates the role of the early postoperative systemic inflammatory response in predicting complications after pancreaticoduodenectomy and aims to establish clinically relevant thresholds for C-Reactive Protein for the prediction of complications. The results of this chapter demonstrate that CRP levels as early as the second postoperative day are associated with complications. While post-operative CRP was useful in the prediction of infective complications, this was the case only in patients who did not develop a post-operative pancreatic fistula. The predictive ability of inflammatory markers for infectious complications was blunted in patients with a pancreatic fistula. Chapter 7 summarises the findings of this thesis, their place in current literature and future directions. The results of this thesis add to the current knowledge regarding the complex pathophysiological abnormalities in patients undergoing pancreaticoduodenectomy, with specific emphasis on the interaction between cardiopulmonary exercise physiology, obstructive jaundice, systemic inflammation and postoperative outcomes. The work presented in this thesis lays the foundations for further studies aimed at improving outcomes after pancreaticoduodenectomy through the development of individualised, goal-directed therapies that are initiated well before this morbid yet necessary operation is performed.

Estimation of whole body muscle, adipose/fat mass, validation in health and during weight loss. Development of prediction equations using MRI as reference method

Background: Body composition is affected by diseases, and affects responses to medical treatments, dosage of medicines, etc., while an abnormal body composition contributes to the causation of many chronic diseases. While we have reliable biochemical tests for certain nutritional parameters of body composition, such as iron or iodine status, and we have harnessed nuclear physics to estimate the body’s content of trace elements, the very basic quantification of body fat content and muscle mass remains highly problematic. Both body fat and muscle mass are vitally important, as they have opposing influences on chronic disease, but they have seldom been estimated as part of population health surveillance. Instead, most national surveys have merely reported BMI and waist, or sometimes the waist/hip ratio; these indices are convenient but do not have any specific biological meaning. Anthropometry offers a practical and inexpensive method for muscle and fat estimation in clinical and epidemiological settings; however, its use is imperfect due to many limitations, such as a shortage of reference data, misuse of terminology, unclear assumptions, and the absence of properly validated anthropometric equations. To date, anthropometric methods are not sensitive enough to detect muscle and fat loss. Aims: The aim of this thesis is to estimate Adipose/fat and muscle mass in health disease and during weight loss through; 1. evaluating and critiquing the literature, to identify the best-published prediction equations for adipose/fat and muscle mass estimation; 2. to derive and validate adipose tissue and muscle mass prediction equations; and 3.to evaluate the prediction equations along with anthropometric indices and the best equations retrieved from the literature in health, metabolic illness and during weight loss. Methods: a Systematic review using Cochrane Review method was used for reviewing muscle mass estimation papers that used MRI as the reference method. Fat mass estimation papers were critically reviewed. Mixed ethnic, age and body mass data that underwent whole body magnetic resonance imaging to quantify adipose tissue and muscle mass (dependent variable) and anthropometry (independent variable) were used in the derivation/validation analysis. Multiple regression and Bland-Altman plot were applied to evaluate the prediction equations. To determine how well the equations identify metabolic illness, English and Scottish health surveys were studied. Statistical analysis using multiple regression and binary logistic regression were applied to assess model fit and associations. Also, populations were divided into quintiles and relative risk was analysed. Finally, the prediction equations were evaluated by applying them to a pilot study of 10 subjects who underwent whole-body MRI, anthropometric measurements and muscle strength before and after weight loss to determine how well the equations identify adipose/fat mass and muscle mass change. Results: The estimation of fat mass has serious problems. Despite advances in technology and science, prediction equations for the estimation of fat mass depend on limited historical reference data and remain dependent upon assumptions that have not yet been properly validated for different population groups. Muscle mass does not have the same conceptual problems; however, its measurement is still problematic and reference data are scarce. The derivation and validation analysis in this thesis was satisfactory, compared to prediction equations in the literature they were similar or even better. Applying the prediction equations in metabolic illness and during weight loss presented an understanding on how well the equations identify metabolic illness showing significant associations with diabetes, hypertension, HbA1c and blood pressure. And moderate to high correlations with MRI-measured adipose tissue and muscle mass before and after weight loss. Conclusion: Adipose tissue mass and to an extent muscle mass can now be estimated for many purposes as population or groups means. However, these equations must not be used for assessing fatness and categorising individuals. Further exploration in different populations and health surveys would be valuable.