Oguz Akbilgic

A novel Hybrid RBF Neural Networks model as a forecaster

We introduce a novel predictive statistical modeling technique called Hybrid Radial Basis Function Neural Networks (HRBF-NN) as a forecaster. HRBF-NN is a flexible forecasting technique that integrates regression trees, ridge regression, with radial basis function (RBF) neural networks (NN). We develop a new computational procedure using model selection based on information-theoretic principles as the fitness function using the genetic algorithm (GA) to carry out subset selection of best predictors. Due to the dynamic and chaotic nature of the underlying stock market process, as is well known, the task of generating economically useful stock market forecasts is difficult, if not impossible. HRBF-NN is well suited for modeling complex non-linear relationships and dependencies between the stock indices. We propose HRBF-NN as our forecaster and a predictive modeling tool to study the daily movements of stock indices. We show numerical examples to determine a predictive relationship between the Istanbul Stock Exchange National 100 Index (ISE100) and seven other international stock market indices. We select the best subset of predictors by minimizing the information complexity (ICOMP) criterion as the fitness function within the GA. Using the best subset of variables we construct out-of-sample forecasts for the ISE100 index to determine the daily directional movements. Our results obtained demonstrate the utility and the flexibility of HRBF-NN as a clever predictive modeling tool for highly dependent and nonlinear data.

Occurrence and origin of methane in groundwater in Alberta (Canada): Gas geochemical and isotopic approaches

To assess potential future impacts on shallow aquifers by leakage of natural gas from unconventional energy resource development it is essential to establish a reliable baseline. Occurrence of methane in shallow groundwater in Alberta between 2006 and 2014 was assessed and was ubiquitous in 186 sampled monitoring wells. Free and dissolved gas sampling and measurement approaches yielded comparable results with low methane concentrations in shallow groundwater, but in 28 samples from 21 wells methane exceeded 10mg/L in dissolved gas and 300,000 ppmv in free gas. Methane concentrations in free and dissolved gas samples were found to increase with well depth and were especially elevated in groundwater obtained from aquifers containing coal seams and shale units. Carbon isotope ratios of methane averaged -69.7 ± 11.1‰ (n=63) in free gas and -65.6 ± 8.9‰ (n=26) in dissolved gas. δ(13)C values were not found to vary with well depth or lithology indicating that methane in Alberta groundwater was derived from a similar source. The low δ(13)C values in concert with average δ(2)HCH4 values of -289 ± 44‰ (n=45) suggest that most methane was of biogenic origin predominantly generated via CO2 reduction. This interpretation is confirmed by dryness parameters typically >500 due to only small amounts of ethane and a lack of propane in most samples. Comparison with mud gas profile carbon isotope data revealed that methane in the investigated shallow groundwater in Alberta is isotopically similar to hydrocarbon gases found in 100-250 meter depths in the WCSB and is currently not sourced from thermogenic hydrocarbon occurrences in deeper portions of the basin. The chemical and isotopic data for methane gas samples obtained from Alberta groundwater provide an excellent baseline against which potential future impact of deeper stray gases on shallow aquifers can be assessed.

Natal cleft deeper in patients with pilonidal sinus: implications for choice of surgical procedure.

PURPOSE: The study was planned to evaluate the depth of natal cleft in patients with pilonidal sinus disease and in healthy persons. METHODS: The study included 50 patients with pilonidal sinus disease and 51 volunteers. Data including body mass index and natal cleft depth were recorded. Natal cleft depth was measured in millimeters by using a caliper instrument. Data were evaluated with the use of the statistical package program (SPSS) with a chi-squared test analysis. P < 0.01 was evaluated as significant. RESULTS: There was no discernable difference in age, occupation, and sex between the groups. The mean natal cleft depth was 27.06 mm in the pilonidal sinus group and 21.07 in the nonpilonidal sinus group. The differences between the two groups were statistically significant (P < 0.01) for natal cleft depth. The mean body mass index was 25.71 in the pilonidal sinus group and 25.28 in the nonpilonidal sinus group. The difference between groups was statistically insignificant for body mass index. CONCLUSIONS: The natal cleft of patients with pilonidal sinus disease is deeper than the natal cleft of members of the volunteer group.

Seasonal variation of the onset of presentations in stage 1 sarcoidosis

Sarcoidosis is a chronic disease with an unknown aetiology. Our aim was to evaluate the pattern of seasonality of stage 1 sarcoidosis subjects who had symptoms by all cases, by age and by both genders. In this study, we used Rogers test for cyclic variation to prove that this seasonal variation was more than chance.

A New Supervised Classification of Credit Approval Data via the Hybridized RBF Neural Network Model Using Information Complexity

In this paper, we introduce a new approach for supervised classification to handle mixed-data (i.e., categorical, binary, and continuous) data structures using a hybrid radial basis function neural networks (HRBF-NN). HRBF-NN supervised classification combines regression trees, ridge regression, and the genetic algorithm (GA) with radial basis function (RBF) neural networks (NN) along with information complexity (ICOMP) criterion as the fitness function to carry out both classification and subset selection of best predictors which discriminate between the classes. In this manner, we reduce the dimensionality of the data and at the same time improve classification accuracy of the fitted predictive model. We apply HRBF-NN supervised classification to a real benchmark credit approval mixed-data set to classify the customers into good/bad classes for credit approval. Our results show the excellent performance of HRBF-NN method in supervised classification tasks.

Social network analysis of scientific collaborations across different subject fields

This paper analyzes the level of scientific collaboration and interaction in different subject fields using a novel social network analysis SNA on a data set provided by the Department of Energy DOE Office of Scientific and Technologic Information OSTI in Oak Ridge, Tennessee. This paper not only determines the level of scientific collaboration between different disciplines, but it also analyzes the trends among the subject fields considered. The results in this paper show clear pattern recognition discovery on the social network model among key subject fields using weighted networks.

Model selection using information criteria under a new estimation method: least squares ratio

In this study, we evaluate several forms of both Akaike-type and Information Complexity (ICOMP)-type information criteria, in the context of selecting an optimal subset least squares ratio (LSR) regression model. Our simulation studies are designed to mimic many characteristics present in real data – heavy tails, multicollinearity, redundant variables, and completely unnecessary variables. Our findings are that LSR in conjunction with one of the ICOMP criteria is very good at selecting the true model. Finally, we apply these methods to the familiar body fat data set.

Sociomarkers and biomarkers: predictive modeling in identifying pediatric asthma patients at risk of hospital revisits

The importance of social components of health has been emphasized both in epidemiology and public health. This paper highlights the significant impact of social components on health outcomes in a novel way. Introducing the concept of sociomarkers, which are measurable indicators of social conditions in which a patient is embedded, we employed a machine learning approach that uses both biomarkers and sociomarkers to identify asthma patients at risk of a hospital revisit after an initial visit with an accuracy of 66%. The analysis has been performed over an integrated dataset consisting of individual-level patient information such as gender, race, insurance type, and age, along with ZIP code-level sociomarkers such as poverty level, blight prevalence, and housing quality. Using this uniquely integrated database, we then compare the traditional biomarker-based risk model and the sociomarker-based risk model. A biomarker-based predictive model yields an accuracy of 65% and the sociomarker-based model predicts with an accuracy of 61%. Without knowing specific symptom-related features, the sociomarker-based model can correctly predict two out of three patients at risk. We systematically show that sociomarkers play an important role in predicting health outcomes at the individual level in pediatric asthma cases. Additionally, by merging multiple data sources with detailed neighborhood-level data, we directly measure the importance of residential conditions for predicting individual health outcomes.

Effects of varying sampling frequency on the analysis of continuous ECG data streams

A myriad of data is produced in intensive care units (ICU) even for short periods of time. This data is frequently used for monitoring patient’s immediate health status, not for real-time analysis because of technical challenges in real-time processing of such massive data. Data storage is also another challenge in making ICU data useful for retrospective studies. Therefore, it is important to know the minimal sampling frequency requirement to develop real-time analysis on ICU data and to develop a data storage plan. In this study, we have applied the Probabilistic Symbolic Pattern Recognition (PSPR) method in Paroxysmal Atrial Fibrillation (PAF) screening problem by analyzing electrocardiogram signals at different sampling frequencies varying from 128 Hz to 8 Hz. Our results show that using PSPR method, we can obtain a classification accuracy of 82.67% in identifying PAF subjects even when the test data is sampled at 8 Hz frequency (73.33% for 128 Hz). This classification accuracy drastically improved to 92% when other descriptive features were used along with PSPR features. The PSPR’s PAF screening ability at low sampling frequency indicates its potential for real-time analysis and wearable embedded computing applications.

Predialysis coronary revascularization and postdialysis mortality

Objectives: Coronary artery bypass grafting (CABG) is associated with better survival than percutaneous coronary intervention (PCI) in patients with mild‐to‐moderate chronic kidney disease (CKD) and end‐stage renal disease (ESRD). However, the optimal strategy for coronary artery revascularization in patients with advanced CKD who transition to ESRD is unclear. Methods: We examined a contemporary national cohort of 971 US veterans with incident ESRD who underwent first CABG or PCI up to 5 years before dialysis initiation. We examined the association of a history of CABG versus PCI with all‐cause mortality following transition to dialysis using Cox proportional hazards models adjusted for time between procedure and dialysis initiation, sociodemographics, comorbidities, and medications. Results: In total, 582 patients underwent CABG and 389 patients underwent PCI. The mean age was 64 ± 8 years, 99% of patients were male, 79% were white, 19% were African American, and 84% had diabetes. The all‐cause post‐dialysis mortality rates after CABG and PCI were 229 per 1000 patient‐years (95% confidence interval [CI], 205‐256) and 311 per 1000 patient years (95% CI, 272‐356), respectively. Compared with PCI, patients who underwent CABG had 34% lower risk of death (multivariable adjusted hazard ratio, 0.66; 95% CI, 0.51‐0.86, P = .002) after initiation of dialysis. Results were similar in all subgroups of patients stratified by age, race, type of intervention, presence/absence of myocardial infarction, congestive heart failure, and diabetes. Conclusions: CABG in patients with advanced CKD was associated lower risk of death after initiation of dialysis compared with PCI.